Regulations (Preambles to Final Rules) - Table of Contents Regulations (Preambles to Final Rules) - Table of Contents
• Record Type: Air Contaminants
• Section: 7
• Title: Section 7 - VII. Feasibility and Regulatory Analyses

VII. Feasibility and Regulatory Analyses

A. Detailed Table of Contents

B. Introduction and Executive Summary


Introduction
Employee Exposures and Benefits
Nonregulatory Alternatives
Technological Feasibility
Costs of Compliance
Economic Impact
Regulatory Flexibility Analysis
Environmental Impact

C. Survey of Affected Industries

SIC 20  Food Products
SIC 21  Tobacco Manufactures
SIC 22  Textile Mill Products
SIC 23  Apparel
SIC 24  Lumber and Wood Products
SIC 25  Furniture and Fixtures
SIC 26  Paper and Allied Products
SIC 27  Printing and publishing
SIC 28  Chemicals
SIC 29  Petroleum Refining
SIC 30  Rubber and Plastics Products
SIC 31  Leather Products
SIC 32  Stone, Clay, Glass, and Concrete
SIC 33  Primary Metal Industries
SIC 34  Fabricated Metal Products
SIC 35  Machinery, Except Electrical
SIC 36  Electrical Machinery
SIC 37  Transportation Equipment
SIC 38  Instruments
SIC 39  Miscellaneous Manufacturing
SIC 40  Railroad Transportation
SIC 42  Motor Freight Transportation and Warehousing
SIC 45  Air Transportation
SIC 47  Transportation Services
SIC 49  Electric, Gas, Sanitary Services
SIC 50  Wholesale Trade
SIC 51  Wholesale Trade
SIC 55  Auto Dealers and Service Stations
SIC 72  Personal Services
SIC 73  Business Services
SIC 75  Auto Repair
SIC 76  Miscellaneous Repair
SIC 80  Health Services

D. Employee Exposures and Benefits

Description of Data Sources Used
Estimates of the Number of Potentially Exposed Employees
Estimates of the Reduction in Illness Cases and Lost Workdays
Estimates of the Number of Employees Potentially at Risk by Type of Hazard
Estimates of the Number of Illness-Related Fatalities Avoided

E. Nonregulatory Alternatives

Introduction
Market Failure
Tort Liability
Workers' Compensation
Standards of Other Organizations
Conclusion

F. Technological Feasibility

Feasibility Determination
Types of Controls
Industry Engineering Controls
Personal Protective Equipment

G. Costs of Compliance

Linking Hazardous Substances by Industry Use and Employee Exposure
Industrial Processes and Control Costs
Compliance Costs by Industry Sector
Per Plant Average Costs

H. Economic Impact, Regulatory Flexibility Analysis, and environmental
Impact Assessment

Economic Impact
Regulatory Flexibility Analysis
Environmental Impact Assessment

I. Supplement 1 - Technical Description of the Sample Survey

J. Supplement 2 - List of Substances the Pose Potentially Hazardous Exposures and Estimates of Numbers of Workers Exposed, by 4-Digit SIC and Substance

K. Supplement 3 - Summary and Comparison of OSHA Site Visit Data

L. Supplement 4 - Tabulated Results of OSHA's 1988 Nationwide Sample Survey

Note. - Supplements 2 through 4 are available in the OSHA Docket Office, Room N2634, U.S. Department of Labor, 200 Constitution Avenue NW., Washington, DC 20210, (202) 523-7894

B. Introduction and executive summary

Introduction

The Occupational Safety and Health Administration (OSHA) is amending its existing air contaminant standards at 29 CFR 1910.1000, Tables Z-1, Z-2, and Z-3. The amendments provide more protective permissible exposure limits (PELS) for about 212 substances, and set new exposure limits for 164 substances currently not regulated by OSHA. The PELs include time-weighted average limits, short term exposure limits, ceiling limits, and, in some cases, skin designations. No changes are being made to the PELs for 52 substances.

Background. Congress enacted the Occupational Safety and Health Act of 1970 (the Act) to achieve several goals, one of which was to protect workers from occupational health hazards. Congress acknowledged the role of occupational exposure in the development of diseases, and addressed in the Act the need to quickly establish minimum health standards to control exposure to hazardous substances. To accomplish Congress' intent, OSHA adopted initial exposure limits for approximately 430 chemicals. Four hundred of these exposure limits were based on the recommendations of the American Conference of Governmental Industrial Hygienists (ACGIH), and 21 were from the American Standards Association (now called the American National Standards Institute). The list of exposure limits was to be updated, improved, and expanded as new knowledge and techniques were developed. To date OSHA has promulgated extensive health standards for only 24 individual chemicals. The rulemaking under consideration here would set exposure limits for about 430 chemicals based on the 1987-88 Threshold Limit Values of the ACGIH, and recommendations of the National Institute for Occupational Safety and Health (NIOSH) of the U. S. Department of Health and Human Services.

The OSH Act requires the Agency to consider the feasibility of proposed and final standards. Executive Order 12291 (46 FR 13197) requires that a regulatory analysis be conducted for any rule having major economic consequences on the national economy, individual industries, geographical regions, or levels of government. The Regulatory Flexibility Act (5 U.S.C. 601 et seq.) similarly requires OSHA to consider the impact of the proposed and final regulations on small entities. This analysis covers these requirements.

Approach. Because this rulemaking involves about 430 chemicals, OSHA has prepared the regulatory impact analysis in two phases. Phase I involved the use of a number of secondary data bases to collect information on the chemicals to be regulated and the industries in which they are used. These data bases provided information on the toxicity and health effects of exposure to the chemicals, and current information on engineering controls in use and emergency response procedures. Two data bases provided information on employee exposures. The 1982 National Occupational Exposure Survey (NOES) was based on a sample of about 4,500 businesses. The data base developed from this survey contains an estimate of the number of persons occupationally exposed to hazardous substances by Standard Industrial Classification (SIC). The second data base was OSHA's Integrated Management Information System (IMIS). The IMIS contains the results of air samples taken since 1979 by OSHA industrial hygienists in the course of compliance inspections. Both the NOES and IMIS data bases provided valuable information on the nature and extent of employee exposures to the substances to be regulated; however, they did not provide complete information on all substances. Supplementary information was obtained from industrial hygienists and engineers. These experts identified exposure controls in use and the number and size of plants most likely to be affected by this rulemaking. These sources have provided OSHA with a substantial body of information on chemical use, exposures and controls.

Phase II of the data collection effort involved a sampling survey of about 5,700 firms in industries where chemical exposures were believed to pose potential problems. The survey, conducted during the first part of 1988, gathered data on chemicals, processes, exposures and controls currently in use. These additional data have permitted OSHA to refine the Phase I preliminary estimates of technical and economic feasibility. In addition, site visits to 90 firms were conducted to verify the data collected on chemicals, processes, controls, and employee exposures.

OSHA has used contractors to assist in these data collection efforts. Three contractors have supplied expert knowledge on the industries affected and the engineering controls needed to reach the proposed exposure levels. These contractors are Kearney/Centaur Division of A.T. Kearney, Meridian Research, and CONSAD. Fu Associates provided data base management support during all phases of this project. Washington Consulting Group designed the sample for the surveyed firms and KCA Research conducted the telephone interviews of these firms.

Employee Exposure and Benefits

Revising OSHA's Z-Table limits for hazardous substances is expected to result in reduced risk of chemically-related disease among exposed employees. Exposure to substances included in the rulemaking has been associated with a variety of adverse health effects, including impairment of organ system functions, mucous membrane irritation, neuropathy, narcosis, allergic sensitization, respiratory disease, cardiovascular disease, and cancer.

Using data from OSHA's IMIS system and information collected from the survey of about 5,700 establishments, OSHA estimates that over 21 million employees are potentially exposed to hazardous substances in the workplace. OSHA also estimates that over four and one-half million employees are currently exposed above the proposed exposure limits for these substances. Table B-1 summarizes OSHA's estimates of the number of workers currently at risk of adverse health effects. OSHA estimates that promulgation of the final rule's exposure limits will result in a potential reduction of over 55,000 work-related illness cases per year, over 23,000 lost-workday illness cases per year, and almost 520,000 lost workdays due to illness per year. OSHA's estimate is that industry compliance with the final rule's exposure limits will result in a reduction of an average of 683 fatalities annually that are caused by exposure to substances that cause cancer, respiratory disease, cardiovascular disease, or liver or kidney disease.


 Table B-1.  Estimated Number of Workers Potentially at Risk of
             Experiencing Adverse Effects, by Type of Adverse Effect(1)
__________________________________________________________________________
                        NO. WORKERS  NO. WORKERS  NO. WORKERS  NO. WORKERS
                        POTENTIALLY  POTENTIALLY      EXPOSED      EXPOSED
                         EXPOSED TO   EXPOSED TO  ABOVE FINAL  ABOVE FINAL
                         SUBSTANCES   SUBSTANCES   LIMITS FOR   LIMITS FOR
                         ASSOCIATED   ASSOCIATED   SUBSTANCES   SUBSTANCES
                        WITH EFFECT  WITH EFFECT      MINIMUM      MAXIMUM
ADVERSE HEALTH EFFECT   MINIMUM EST. MAXIMUM EST.    ESTIMATE     ESTIMATE
__________________________________________________________________________
PHYSICAL IRRITANT
EFFECTS                   3,375,472    3,889,261      222,191      222,191
ODOR EFFECTS                519,318      521,938        3,597        3,597
SYSTEMIC TOXICITY         4,305,578    5,038,573      457,104      490,282
MUCOUS MEMBRANE
  IRRITATION             10,730,691   14,906,090      789,461    1,141,133
METABOLIC INTERFERENCES   4,015,702    4,205,530    1,233,413    1,241,564
LIVER/KIDNEY DISEASE      3,292,993    3,806,226      536,945      546,429
OCULAR DISTURBANCES       2,482,449    2,569,950       83,272      110,560
RESPIRATORY DISEASE       4,231,235    4,782,280    1,405,501    1,568,579
CARDIOVASCULAR DISEASE      166,077      166,868       44,403       44,403
NEUROPATHY                2,212,358    2,463,583      319,974      401,576
NARCOSIS                  6,966,024   10,520,982      941,472    1,073,717
CANCER                    1,712,799    1,851,342      465,013      528,650
ALLERGIC SENSITIZATION    2,545,551    2,648,973      305,955      305,955
__________________________________________________________________________
  Footnote(1) Double counting of employees simultaneously exposed to more
than one substance in different adverse health effects categories prevents
the summation of workers exposed to all adverse health effects in this
table.

Nonregulatory Alternatives

OSHA believes that there are no nonregulatory alternatives that adequately protect most workers from the adverse health effects associated with exposure to the chemicals under consideration. OSHA believes that the tort liability laws and Workers' Compensation do not provide adequate worker protection due to market imperfections. Some employers have not complied with the standards recommended by professional organizations. The deleterious health effects resulting from continued high levels of exposure to hazardous substances require a regulatory solution, and the OSH Act requires the Agency to protect workers' health.

Technological Feasibility

Consistent with OSHA regulations and policy, engineering controls and work practices are preferred over personal protective equipment to control employee exposures to airborne contaminants.

Engineering controls involve the use of local exhaust ventilation, general ventilation, isolation of the worker and enclosure of the source of emissions, process modifications, equipment modifications, and substitution of non-hazardous or less hazardous chemicals. These methods may be used alone or in combination, depending upon the industrial processes involved. These controls are widely used and will effectively control exposures either by themselves, or coupled with changes in work practices.

Perhaps the most widely used technique for controlling chemical exposure is the use of ventilation. General ventilation uses the movement of air within the general work space to displace or dilute the contaminant with fresh outside air. General ventilation may not be the preferred control method, however, due to the large volumes of air movement required. Local exhaust ventilation uses much smaller volumes of air and controls emissions at the point or source from which contaminants are generated.

Isolation involves placing a physical barrier between the hazardous operation and the worker. Many modern, automated manufacturing processes are now fully enclosed in ventilated cabinets. The effectiveness of such a control technique depends on the frequency with which the workers have to enter the enclosure during normal operations. In other situations, the worker, rather than the process or machine, can be placed in an enclosure having a controlled atmosphere. Many processes which involve potential chemical exposures are operated remotely by operators from air-conditioned booths isolated from the hazardous materials.

Substitution refers to the replacement of a toxic chemical in a particular process or work area with another, less toxic or non-toxic product. Properly applied, substitution can be a very effective control technique. However, care must be taken to ensure that the proposed substitute performs in a similar manner to the product being replaced. In addition, it is essential that the substitute be carefully evaluated to ensure that in controlling one hazard, another different hazard is not inadvertently introduced. The substitute must also be compatible with existing manufacturing equipment and processes.

The success of these engineering control techniques will depend on the physical properties of the chemicals and emissions encountered (boiling point, vapor pressure, etc.) and the process operating conditions. In some cases, particularly with cleaning solvents, substitution may provide the quickest and most effective means of reducing exposure. In other situations, a major effort may be required to alter processes or install or expand local or general dilution ventilation.

OSHA has found that engineering controls and improved work practices are available to reduce exposure levels to the new levels in almost all circumstances. Standard controls have been adapted in numerous situations to solve situation-specific problems in all of the industry sectors affected. Detailed industry-specific illustrations of this point are presented in the Technological Feasibility Chapter of this Feasibility and Regulatory Analysis. OSHA does recognize, however, that in some circumstances, respiratory protection may be necessary to complement engineering controls and that respiratory protection may also be necessary to achieve compliance in some specific operations in some industries.

Costs Of Compliance

Costs of compliance with the proposed rulemaking would result from industry actions to lower workers' chemical exposures to the levels promulgated in the final rule. The 1988 sample survey of almost 5,700 firms was drawn from a universe of over one million firms potentially affected by the rule. Table B-5 at the end of this section presents a list of industries included in the analysis.

Survey respondents verified the number of work stations and workers related to each process, the process location and configuration, the controls already in place, and potential chemical exposures above new proposed levels. Process controls in place were compared to a list of control designs needed to limit exposures to the new, lower levels. Where the required controls were not reported to be in place, a compliance cost per work station was assigned. Process control costs were summed per establishment and certain maintenance workers were assigned a respirator cost. Costs for the surveyed establishments were then weighted (by SIC and size) to represent compliance costs for the universe of affected plants.

The survey found that over 500,000 establishments (of the 1,101,600 establishments covered by the survey) reported using the chemicals being regulated. Of this number, 131,005 would incur some costs to comply with the new limits. The total estimated annualized capital plus annual operating costs are $787.98 million. Table B-2 presents the annual cost by industry sector and the average per plant annual cost for large and small (fewer than 20 employees) plants.

Among all industry sectors the average annual cost per affected establishment will be $6,000.


TABLE B-2. Average Per Plant Annual Costs and Numbers of Affected
           Plants (a)

    Note: Due to its width, this Table has been divided; see continuation
          for additional columns.
__________________________________________________________________________
                                                           # OF
                                              TOTAL #  AFFECTED          %
SIC (b)  SIC DESCRIPTION       ANNUAL COST  OF PLANTS    PLANTS   AFFECTED
__________________________________________________________________________
20    FOOD PROD. (c)           $33,493,100     29,000     4,932     16.98%
21    TOBACCO (c)                  $19,700        200         3      1.39%
22    TEXT. MILL (c)           $29,478,400     11,000     2,765     25.08%
23    APPAREL PROD. (c)        $31,744,200     30,000     6,179     20.57%
24    LUMBER & WOOD            $56,720,800     27,100    18,427     68.00%
25    FURNITURE                $21,075,800     12,700     5,062     40.00%
26    PAPER PROD.              $30,998,700      7,000     3,518     50.00%
27    PRINTING & PUB.          $33,754,500     60,300     3,597      6.88%
28    CHEMICAL PROD.           $35,454,700     16,400     3,007     18.31%
29    PETRO. REFINING          $23,686,000      2,300       306     13.25%
30    RUBBER & PLASTICS       $111,093,400     15,100     3,562     26.22%
31    LEATHER PROD.             $2,414,700      2,300       300     13.46%
32    STONE & CLAY             $22,457,800     15,900     3,267     22.80%
33    PRIM. METAL              $70,957,600      8,000     2,411     30.03%
34    FAB. METALS              $39,419,700     37,300     4,597     14.50%
35    MACHINERY                $45,206,600     64,400     6,801     10.56%
36    ELEC. MACH.              $20,667,500     21,600     2,359     10.92%
37    TRANS. EQUIP.            $49,792,400     13,600     4,979     36.56%
38    INSTRUMENTS               $9,633,500     12,000     1,289     10.74%
39    MISC. MANUF.             $15,842,600     25,300     2,649     10.47%
40    R.R. TRANS.               $1,083,400        400        93     20.86%
45    AIR TRANS.                $3,740,500      5,500       320      5.79%
47    TRANS. SERV.              $3,789,400     26,200       324      1.24%
49    ELEC. GAS. SAN.          $38,009,300     15,800     3,485     22.24%
50    WHOLESALE TRADE           $2,995,300      5,800       801     13.78%
51    WHOLESALE, NON-DUR       $14,215,800     33,600     4,436     13.22%
55    AUTO DEALERS             $13,550,500    165,800    24,847     14.99%
72    PERSONAL SRV.            $10,872,100     95,500     5,217      5.47%
73    BUSINESS SRV.             $2,422,100     12,100       800      6.61%
75    AUTO REPAIR               $6,143,500     91,500     8,351      9.13%
76    MISC. REPAIR SRV.         $2,809,900     15,100     1,163     11.56%
80    HEALTH SERV. (c)          $4,439,400    222,800     1,158      0.52%
__________________________________________________________________________
TOTAL                         $787,982,900  1,101,600   131,005     11.89%

Table B-2  Average per Plant Annual Costs and Numbers of Affected Plants
           (a)(Continuation)
__________________________________________________________________________
                                   AVERAGE         AVERAGE         AVERAGE
                                  COST PER  COST PER LARGE  COST PER SMALL
SIC (b)  SIC DESCRIPTION      AFFTED PLANT    AFFTED PLANT    AFFTED PLANT
__________________________________________________________________________
20    FOOD PROD. (c)                $6,800         $13,000          $3,600
21    TOBACCO (c)                   $6,600          $6,600              $0
22    TEXT. MILL (c)               $10,700         $21,400          $3,700
23    APPAREL PROD. (c)             $5,100         $11,500          $2,000
24    LUMBER & WOOD                 $3,100          $4,200          $2,700
25    FURNITURE                     $4,200         $12,400          $1,800
26    PAPER PROD.                   $8,800         $15,200            $800
27    PRINTING & PUB.               $9,400          $6,200         $10,600
28    CHEMICAL PROD.               $11,800         $16,200          $5,400
29    PETRO. REFINING              $77,400        $106,600            $700
30    RUBBER & PLASTICS            $31,200         $27,000         $35,100
31    LEATHER PROD.                 $8,000         $10,400          $6,400
32    STONE & CLAY                  $6,900         $12,200          $3,400
33    PRIM. METAL                  $29,400         $41,900          $6,200
34    FAB. METALS                   $8,600         $15,800          $3,800
35    MACHINERY                     $7,800         $14,600          $3,000
36    ELEC. MACH.                   $7,800         $14,500          $3,000
37    TRANS. EQUIP.                $10,000         $11,800          $8,800
38    INSTRUMENTS                   $7,800         $14,500          $3,000
39    MISC. MANUF.                  $7,800         $14,600          $3,000
40    R.R. TRANS.                  $11.700         $11,700              $0
45    AIR TRANS.                   $11,700         $11,700              $0
47    TRANS. SERV.                 $11,700         $11,700              $0
49    ELEC. GAS. SAN.              $10,900         $17,000          $3,600
50    WHOLESALE TRADE               $3,400          $6,200          $2,900
51    WHOLESALE, NON-DUR            $3,400          $6,200          $2,900
55    AUTO DEALERS                    $360          $2,000            $300
72    PERSONAL SRV.                 $2,200          $6,000          $1,000
73    BUSINESS SRV.                 $2,200          $8,300          $1,500
75    AUTO REPAIR                     $600          $3,500            $300
76    MISC. REPAIR SRV.             $2,400         $12,400          $2,100
80    HEALTH SERV. (c)              $3,800         $12,500          $2,100
__________________________________________________________________________
TOTAL                               $6,000         $13,000          $3,100
__________________________________________________________________________
Source:  U.S. Department of Labor, Occupational Safety and Health
Administration, Office of Regulatory Analysis.

  Footnote(a) Costs were calculated by annualizing the capital cost over
the projected life of the equipment (10 years) using a 10 percent cost of
capital and adding an annual operating and maintenance cost estimated at
10 percent of the capital cost.
  Footnote(b) Industry sectors not identified in this table include
industries with no major cost impact expected, the construction industry,
which will be the subject of a separate regulatory analysis, and
industries such as mining, over which OSHA has no jurisdiction.
  Footnote(c) Costs in these sectors were based on expert judgement and
secondary data collection.

Economic Impact

OSHA prepared two estimates of the economic effects of this regulation on potentially affected firms. The two estimates were based upon No Cost-Passthrough ("worst case") and Total Cost-Passthrough ("best case") scenarios.

In the first scenario it was assumed that all compliance costs would be absorbed by firms in the form of reduced profits. Table B-3 contains a summary of this "worst case" analysis. Under this scenario, the estimated average percent reduction in profits for all affected firms was less than one percent. The estimated reduction in profit of 2.3 percent for SIC 30 Rubber and Plastics was the highest among all industries.


Table B-3.  ECONOMIC EFFECTS: NO-COST PASSTHROUGH SCENARIO(1)

    Note: Due to its width, this Table has been divided; see continuation
          for additional columns.
__________________________________________________________________________
                                                    R.O.R.
                             Annual          Total      on
                           Costs(2)       Sales(3)   Sales         Pre-Reg
SIC   Industry          ($millions)   ($ millions)  (%)(4)   Profits ($ m)
__________________________________________________________________________
20    FOOD PROD.              33.49     353,780.38     1.9        8,008.04
21    TOBACCO                  0.02      74,030.13     5.3        3,923.60
22    TEXT. MILL              29.48      60,735.22     2.7        1,765.42
23    APPAREL PROD.           31.74      74,474.65     2.8        1,813.22
24    LUMBER & WOOD           56.72      57,994.48     3.9        1,974.51
25    FURNITURE               21.08      37,648.27     3.5        1,411.02
26    PAPER PROD.             31.00     103,694.14     3.7        3,778.20
27    PRINTING & PUB.         33.75     134,830.21     4.8        6,471.85
28    CHEMICAL PROD.          35.45     272,759.67     3.7       11,738.80
29    PETRO. REFINING         23.69     196,400.57     2.7        4,964.85
30    RUBBER & PLASTICS      111.09      86,538.58     4.3        3,423.75
31    LEATHER PROD.            2.41      15,449.56     2.6          401.69
32    STONE & CLAY            22.46      46,094.04     4.1        1,954.99
33    PRIM. METAL             70.96     112,564.26     3.3        3,714.62
34    FAB. METALS             39.42     150,146.41     4.0        6,005.86
35    MACHINERY               45.21     345,144.89     5.1       17,602.39
36    ELEC. MACH.             20.67     245,982.70     5.0       12,299.14
37    TRANS. EQUIP.           49.79     365,427.20     3.9       14,520.25
38    INSTRUMENTS              9.63      83,359.57     4.9        3,373.26
39    MISC. MANUF.            15.84      41,870.30     4.4        1,788.56
40    R.R. TRANS.              1.08      43,869.14    10.0        3,969.62
45    AIR TRANS.               3.74     109,538.08     3.6        3,251.40
47    TRANS. SERV.             3.79      12,254.96     2.7          582.18
49    ELEC. GAS. SAN.         38.01     300,254.83     7.0       21,017.84
50    WHOLESALE TRADE(5)       3.00      13,853.52     2.0          277.07
51    WHOLESALE, NON-DUR      14.22     113,848.20     1.5        1,726.26
55    AUTO DEALERS            13.55     341,574.50     1.9        6,489.92
72    PERSONAL SRV.           10.87      24,270.74     7.3        1,771.76
73    BUSINESS SRV.            2.42      22,165.94     6.6        1,462.95
75    AUTO REPAIR              6.14      45,750.92     5.1        2,492.19
76    MISC. REPAIR SRV.        2.81       2,665.52     5.5          146.60
80    HEALTH SERV. (c)         4.44     170,234.25     4.5        7,807.72
__________________________________________________________________________

Table B-3  Economic Effects: No-Cost Passthrough
           Scenario(1) - Continued
____________________________________________________________
                                   Post-Reg        % Change
SIC   Industry                Profits ($ m)      in Profits
____________________________________________________________
20    FOOD PROD.                   7,896.29         - 0.2715
21    TOBACCO                      3,923.59         - 0.0003
22    TEXT. MILL                   1,747.59         - 1.0100
23    APPAREL PROD.                1,793.56         - 1.0845
24    LUMBER & WOOD                1,931.92         - 2.1574
25    FURNITURE                    1,398.82         - 0.8645
26    PAPER PROD.                  3,761.12         - 0.4519
27    PRINTING & PUB.              6,444.77         - 0.4185
28    CHEMICAL PROD.              11,717.79         - 0.1790
29    PETRO. REFINING              4,952.04         - 0.2579
30    RUBBER & PLASTICS            3,343.76         - 2.3361
31    LEATHER PROD.                  400.03         - 0.4127
32    STONE & CLAY                 1,940.97         - 0.7170
33    PRIM. METAL                  3,674.83         - 1.0712
34    FAB. METALS                  5,981.33         - 0.4084
35    MACHINERY                   17,573.57         - 0.1637
36    ELEC. MACH.                 12,286.86         - 0.0998
37    TRANS. EQUIP.               14,485.24         - 0.2411
38    INSTRUMENTS                  3,367.32         - 0.1763
39    MISC. MANUF.                 1,778.14         - 0.5825
40    R.R. TRANS.                  3,964.04         - 0.0147
45    AIR TRANS.                   3,249.38         - 0.0621
47    TRANS. SERV.                   580.13         - 0.3515
49    ELEC. GAS. SAN.             20,994.71         - 0.1100
50    WHOLESALE TRADE(5)             274.56         - 0.9048
51    WHOLESALE, NON-DUR           1,718.59         - 0.4447
55    AUTO DEALERS                 6,480.69         - 0.1422
72    PERSONAL SRV.                1,763.60         - 0.4606
73    BUSINESS SRV.                1,460.94         - 0.1375
75    AUTO REPAIR                  2,488.29         - 0.1563
76    MISC. REPAIR SRV.              144.36         - 1.5298
80    HEALTH SERV.                 7,804.54         - 0.0406
____________________________________________________________
Source:  U.S. Department of Labor, Occupational Safety and Health
         Administration, Office of Regulatory Analysis.

  Footnote(1) All values in 1985 dollars.
  Footnote(2) Reproduced from table G-1.
  Footnote(3) Dun and Bradstreet, Dun's Marketing Identifiers (DMI)
Database.
  Footnote(4) Rate of Return on Sales, Dun and Bradstreet, Industry Norms
Database.
  Footnote(5) Consists of SIC 5093 (scrap and waste materials) only.

In the second scenario it was assumed that all compliance costs would be passed on to the consumer in the form of higher prices. The potential price increase for an industry sector at the two-digit SIC level was estimated by dividing the sector's compliance cost by its total sales. In this scenario, there would be little impact on market prices; none of the estimated price increases exceeded one-half of one percent (see Table B-4).


TABLE B-4.  ECONOMIC EFFECTS: TOTAL-COST PASSTHROUGH
_____________________________________________________________________
                                                             Costs as
                          Annual Costs       Total Sales    a Percent
SIC    Industry           ($ millions)      ($ millions)     of Sales
_____________________________________________________________________
20    FOOD PROD.                 33.49        353,780.38       0.0095
21    TOBACCO                     0.02         74,030.13       0.0000
22    TEXT. MILL                 29.48         60,735.22       0.0485
23    APPAREL PROD.              31.74         74,474.65       0.0426
24    LUMBER & WOOD              56.63         57,994.48       0.0978
25    FURNITURE                  26.28         37,648.28       0.0560
26    PAPER PROD.                33.00        103,694.14       0.0299
27    PRINTING & PUB.            34.39        134,830.21       0.0250
28    CHEMICAL PROD.             38.87        272,759.67       0.0130
29    PETRO. REFINING            23.91        196,400.57       0.0121
30    RUBBER & PLASTICS         121.93         86,538.58       0.1284
31    LEATHER PROD.               2.66         15,449.56       0.0156
32    STONE & CLAY               25.83         46,094.04       0.0487
33    PRIM. METAL                78.24        112,564.26       0.0630
34    FAB. METALS                53.51        150,146.41       0.0263
35    MACHINERY                  50.00        345,144.89       0.0131
36    ELEC. MACH.                23.30        245,982.70       0.0084
37    TRANS. EQUIP.              49.79        365,427.20       0.0136
38    INSTRUMENTS                10.75         83,359.57       0.0116
39    MISC. MANUF.               17.29         41,870.30       0.0378
40    R.R. TRANS.                 1.09         43,869.14       0.0025
45    AIR TRANS.                  3.76        109,538.08       0.0034
47    TRANS. SERV.                3.81         12,254.96       0.0309
49    ELEC. GAS. SAN.            37.83        300,254.83       0.0127
50    WHOLESALE TRADE(1)          3.13         13,853.52       0.0216
51    WHOLESALE, NON-DUR         14.80        113,848.20       0.0125
55    AUTO DEALERS               22.72        341,574.50       0.0040
72    PERSONAL SRV.              10.87         24,270.74       0.0448
73    BUSINESS SRV.               2.42         22,165.94       0.0109
75    AUTO REPAIR                10.25         45,750.92       0.0134
76    MISC. REPAIR SRV.           4.86          2,665.52       0.1054
80    HEALTH SERV. (c)            4.44        170,234.25       0.0026
_____________________________________________________________________
Source:  U.S. Department of Labor, Occupational Safety and Health
         Administration, Office of Regulatory Analysis

  Footnote(1) Consists of SIC 5093 (scrap and waste materials) only.

Based on this analysis, OSHA concludes that the final standard is economically feasible for each sector. The impact on prices is slight and, even in the worst cases, the reductions in profitability are small.

Regulatory Flexibility Analysis

In accordance with the Regulatory Flexibility Act (P.L. 96-353, 94 Stat. 1664) [5 U.S.C. 601 et seq.], OSHA has made a preliminary assessment of how this rulemaking will affect large and small establishments. The results of this preliminary assessment indicate that some small establishments may experience some adverse impact. The smaller profit margins of some small establishments may make it difficult for them to absorb increases in compliance costs. An important ameliorating factor for each affected firm will be its ability to pass through additional costs to the consumer. The ability of individual firms to do this will be dependent upon product demand elasticities. It is expected that all impacted firms will be able to pass through some portion of their increased costs.

Environmental Impact

The standard has been reviewed in accordance with the requirements of the National Enviromental Policy Act of 1969 (NEPA), the Council on Environmental Quality NEPA regulations, and the Department of Labor's NEPA compliance procedures and is not anticipated to have a significant impact on the external environment.


                            TABLE B-5
                SIC GROUPS COVERED IN THE OSHA ANALYSIS
_________________________________________________________________________
Division D.  Manufacturing

              Major Group 20. Food and kindred products
              Major Group 21. Tobacco manufactures
              Major Group 22. Textile mill products
              Major Group 23. Apparel and other finished
                              products, made from fabrics
                              and similar materials
              Major Group 24. Lumber and wood products,
                              except furniture
              Major Group 25. Furniture
              Major Group 26. Paper and allied products
              Major Group 27. Printing, publishing, and
                              allied industries
              Major Group 28. Chemicals and allied products
              Major Group 29. Petroleum refining and
                              related industries
              Major Group 30. Rubber and miscellaneous
                              plastics products
              Major Group 31. Leather and leather products
              Major Group 32. Stone, clay, glass, and
                              concrete products
              Major Group 33. Primary metal industries
              Major Group 34. Fabricated metal products,
                              except machinery and
                              transportation equipment
              Major Group 35. Machinery, except electrical
              Major Group 36. Electrical and electronic
                              machinery, equipment, and
                              supplies
              Major Group 37. Transportation equipment
              Major Group 38. Measuring, analyzing, and
                              controlling instruments;
                              photographic, medical and
                              optical goods; watches and
                              clocks
              Major Group 39. Miscellaneous manufacturing
                              industries

Division E.  Transportation, Communications, Electric, Gas, and Sanitary
              Services

              Major Group 40. Railroad transportation
              Major Group 45. Transportation by air
              Major Group 47. Transportation services
              Major Group 49. Electric, gas, and sanitary
                              services

Division F.  Wholesale Trade

              Major Group 50. Wholesale trade - durable goods
              Major Group 51. Wholesale trade - nondurable goods

Division G.  Retail Trade

              Major Group 55. Automotive dealers and gasoline service
                              stations

Division I.  Services
              Major Group 72. Personal services
              Major Group 73. Business services
              Major Group 75. Automotive repair, services, and garages
              Major Group 76. Miscellaneous repair services
              Major Group 80. Health services
__________________________________________________________________________
Source:  U.S. Department of Labor, OSHA, Office of Regulatory Analysis, as
         derived from Standard Industrial Classification Manual 1972,
         Executive Office of the President -- Office of Management and
         Budget.

The listing excludes the construction industry (SICs 15, 16, and 17) which will be the subject of a separate regulatory analysis.

C. Survey Of Affected Industries

Chemicals and other hazardous substances are present to some degree in all industries. However, some industry sectors use chemicals more extensively than others and have controls in place which do not always reduce workers' exposures below permissible exposure levels. This chapter presents an overview of those industries which OSHA believes may experience costs and benefits as a result of this rulemaking. In order to estimate and quantify the potential impact of the rule, a sample survey of about 5,700 establishments was conducted during the first part of 1988. The results of the survey provided the basis for the cost and benefit estimates presented in this Feasibility and Regulatory Analyses.

Table C-1 at the end of this chapter shows establishment and employment data for the industries where OSHA expects costs and benefits. In order to determine which industries to include in the sample survey, OSHA relied primarily on two data sources: 1) the NIOSH National Occupational Exposure Survey (NOES) of 1982 and supplementary information from the NIOSH 1972 survey; and 2) data in the OSHA Integrated Management Information System (IMIS). The 1982 NOES database contains a sample of the number of persons exposed by substance and industry from almost 4,500 businesses in 98 different geographic areas in the United States. OSHA's IMIS contains the results of exposure samples taken since 1979 by industrial hygienists during the course of compliance inspections. Using these two databases, industries which are likely to use the substances in this rulemaking at levels which might exceed the proposed exposure limits were identified.

As a check on this list of industries, OSHA contracted with about one dozen industrial hygienists and chemical engineers to review the list. Based on their professional knowledge, these experts verified the industries with potential exposure problems. The final list of industries selected for the sample survey included over 30 two-, three-, and four-digit SICs where it is believed that chemical exposures potentially exceed the new or revised levels.

Industry sectors not included in the survey are those where OSHA believes there is little potential chemical exposure or where existing exposures are well controlled. Industries which were not surveyed for these reasons included finance, real estate, insurance and most service and retail trade sectors. The construction industry was also excluded and will be the subject of a separate rulemaking action. Industries such as mining and certain transportation sectors were not included since other agencies have safety and health enforcement jurisdiction. Certain industry sectors including textile, apparel, food and tobacco products are expected to incur some costs as a result of this rulemaking, but these were not included in the sample survey. The reasons for not including these sectors in the survey were restraints on the sample size, relatively low hazardous substance exposure levels, and the availability of adequate information on the engineering controls currently in use in these industries.

Industrial hygienists and engineers under contract to OSHA also identified the processes used in the industries surveyed, and the chemicals used in those processes. Expected levels of exposure and the number of employees potentially exposed were estimated. The list of processes and chemicals determined to be in common use in each industry sector was subsequently verified in the sample survey.

Establishments to be surveyed were selected based on a statistical sample of all establishments in the surveyed industry sectors. For each SIC, establishments were selected from four size categories:

(a) 0-19 employees (b) 20-99 employees (c) 100-249 employees (d) 250 or more employees This permitted analysis of the effects of the rulemaking by establishment employment size.

About 5,000 completed responses were required to obtain statistically valid results. The field survey was conducted by KCA Research using Computer-Assisted Telephone Interviewing (CATI). Trained interviewers requested data from each establishment regarding production employment, chemical usage, and exposure guidelines in use. Respondents were asked to verify the presence or absence of chemicals and processes believed to be found in establishments in their industry, and were asked to volunteer information on other chemicals not included on the interviewers' "prompt" list of chemicals in use. For each chemical present, the respondent was asked about amounts used, employee exposure levels, and processes where used. For each process, the respondent was asked questions concerning its configuration, frequency of use, and the types of controls and personal protective equipment in use. This information was used to develop the estimates of costs and benefits presented in this RIA.

Supplement 1 contains a technical summary of the survey and Supplement 4 contains tabulations of the survey results. (Survey results include some responses from SIC 44 - Water Transportation and SIC 46 - Pipelines. These were included prior to a determination that the SICs included industries not within the scope of this rulemaking, or where other agencies have jurisdiction.) The results generally corroborated the preliminary assessments of potential industry exposures and overexposure to chemicals and provided a general picture of workers' exposure in these industries. In the sample of about 5,700 firms, over one-half reported chemicals being used in the workplace. Most of the firms which reported no chemical usage were small administrative or distribution units of multi-plant companies. Among the firms surveyed which use chemicals, almost one-third use specific exposure standards as targets for maintaining workers' exposure. The OSHA PELs are used by 59 percent of firms with specific exposure standards, ACGIH TLVs are used by 22 percent and the NIOSH RELs by one percent. Table C-2 shows the distribution of adopted exposure standards by surveyed industry groups. Over one-third of all firms reported that they have a hazard communication training program; however, less than one-half of the firms using chemicals reported having a hazard communication program (see Table C-3).


Table C-2 - NUMBER OF FIRMS WITH CHEMICALS USING SPECIFIC EXPOSURE
            STANDARDS
__________________________________________________________________________
                                                            DON'T
                       OSHA  NIOSH   ACGIH                   KNOW
CELL                  PEL'S  REL'S   TLV'S  OTHER   NONE  REFUSED    TOTAL
__________________________________________________________________________
CELL 1 SIC's 243      1,133    215     307  1,109   6,238      397  10,137
CELL 2 SIC's 245        202     16      16     35     437       20     754
CELL 3 SIC's 249        932      7      32    326   1,752      169   3,302
CELL 4 SIC's 25       2,266      9     465  1,477   3,648        0   8,361
CELL 5 SIC's 26       1,381     11     176     51   1,273      256   3,490
CELL 6 SIC's 27       4,980    362   2,297  7,148  24,008    3,438  45,757
CELL 7 SIC's 281        705     32     245     35     268        2   1,298
CELL 8 SIC's 282        393     39     190     53     123        0     830
CELL 9 SIC's 283        297     24     149     86     282       20     913
CELL 10 SIC's 284       653     13     272     23      89        0   1,285
CELL 11 SIC's 285       720     36     240    135     112       24   1,377
CELL 12 SIC's 286       318     10     210     13     128        3     742
CELL 13 SIC's 287       105      8     100    143     568        0     929
CELL 14 SIC's 289       580    147     525      2     680        2   1,970
CELL 15 SIC's 291       220      2      43     19      43        0     347
CELL 16 SIC's 295       153     12      27     11     163       26     417
CELL 17 SIC's 299       107      8      65      1      97        6     285
CELL 18 SIC's 307     2,155      0   1,593    291   2,117      108   6,280
CELL 19 SIC's 301-306   281     16   1,575     27     202        3   2,128
CELL 20 SIC's 3111       32      2       8      0      52        0      94
CELL 21 SIC's 313-319   253      0      25     25     599      186   1,306
CELL 22 SIC's 32      4,028      0     714     11   5,363       92  11,308
CELL 23 SIC's 33      2,999    108     641    151   1,152      133   5,297
CELL 24 SIC's 34      8,072     31   2,162    242   6,104    1,575  19,386
CELL 25 SIC's 35,36,
  38,39              18,560    588   4,560  2,078  25,724    2,637  57,492
CELL 26 SIC's 40,44,
  45,47                 293      0     207      0     213       72     815
CELL 27 SIC's 46        170      0      28      0      19        1     226
CELL 28 SIC's 49      2,513     19     718    130   1,609      458   5,805
CELL 29 SIC's 5093,
  5153,5161,5191,
  5198                4,098     60   2,199    769   4,377      338  12,024
CELL 31 SIC's 55,75  39,349      0  29,674     82  58,039   28,443 169,312
CELL 32 SIC's 72,73  18,615      0   5,368  5,251  13,208    8,713  52,189
CELL 33 SIC's 7641,
  7692                2,082      0     708    242   2,872    2,564   8,963
CELL 34 SIC's 80     40,858    117   4,768    510  34,671    8,077  94,202
CELL 99 SIC's 37      6,924     14     411      0   3,900        0  11,263
                     _______ _____ _______ ______ _______  _______ _______
TOTAL                166,425 1,907  60,720 20,476 200,131   57,762 540,285


Table C-3. FIRMS WITH CHEMICALS REPORTING HAZARD COMMUNICATIONS TRAINING
           (Q61) (WEIGHTED)
___________________________________________________________________
                                           NO       DON'T
                        TRAINING     TRAINING        KNOW
CELL                     PROGRAM      PROGRAM     REFUSED     TOTAL
___________________________________________________________________
CELL 1 SIC'S 243           4,478        5,310         349    10,137
CELL 2 SIC'S 245             651           76          28       754
CELL 3 SIC'S 249           1,202        2,080          20     3,302
CELL 4 SIC'S 25            3,110        5,246           5     8,361
CELL 5 SIC'S 26            2,906          374         210     3,490
CELL 6 SIC'S 27           19,543       26,214           0    45,757
CELL 7 SIC'S 281           1,193           81          24     1,298
CELL 8 SIC'S 282             673          156           0       830
CELL 9 SIC'S 283             752          155           5       913
CELL 10 SIC'S 284          1,044          232           8     1,285
CELL 11 SIC'S 285          1,175          177          24     1,377
CELL 12 SIC'S 286            595          136          11       742
CELL 13 SIC'S 287            925            3           2       929
CELL 14 SIC'S 289          1,802          166           2     1,970
CELL 15 SIC'S 291            326           16           4       347
CELL 16 SIC'S 295            317          100           0       417
CELL 17 SIC'S 299            205           74           6       285
CELL 18 SIC'S 307          4,578        1,701           0     6,280
CELL 19 SIC'S 301-306      2,069           55           3     2,128
CELL 20 SIC'S 3111            68           26           0        94
CELL 21 SIC'S 313-319        418          703         186     1,306
CELL 22 SIC'S 32           6,322        4,894          92    11,308
CELL 23 SIC'S 33           4,364          923          10     5,297
CELL 24 SIC'S 34          12,136        7,226          24    19,386
CELL 25 SIC'S 35,36,
  38,39                   29,929       25,527       2,037    57,492
CELL 26 SIC'S 40,44,
  45,47                      793           22           0       815
CELL 27 SIC'S 46             226            0           0       226
CELL 28 SIC'S 49           4,193        1,612           0     5,805
CELL 29 SIC'S 5093,
  5153,5161,5191,5198      8,231        3,789           4    12,024
CELL 31 SIC'S 55,75       57,277      112,035           0   169,312
CELL 32 SIC'S 72,73       25,332       26,857           0    52,189
CELL 33 SIC'S 7641,
  7692                     2,791        5,955         217     8,963
CELL 34 SIC'S 80          23,583       66,579       4,040    94,202
CELL 99 SIC'S 37           7,631        3,633           0    11,263
                         _______      _______       _____   _______
TOTAL                    230,839      302,135       7,311   540,285
-------------------------------------------------------------------

Some commenters objected to the use of the telephone survey method in lieu of written responses [see, for example, Exs. 3-750, 3-877 and 3-747]. They stated that the questions were too diverse and complex to be answered by a single person, and that the use of the CATI techniques necessitated simplified responses to questions.

As preparation for the survey, OSHA sent a letter to each potential respondent approximately two weeks in advance of the initial phone contact. The letter described the nature of the project, the topics to be covered by the survey and a response card to be returned to the survey contractor listing the name of the person best able to answer the questions. When requested, a copy of the survey questionnaire was provided. With this advance preparation, OSHA believes that respondents were able to accurately and completely answer the questions. While all firms were encouraged to complete the survey over the telephone so that the responses could be entered on the computer during the interview, some firms refused to do so and also failed to return the written survey forms. Overall, the survey achieved a 60 percent completion ratio (the ratio of completed questionnaires to total sample cases drawn, both in and out of scope). OSHA believes that the use of the CATI technique greatly improved the response rate to the survey. Previous OSHA surveys have had completion ratios as low as 30 percent.

To reduce the burden on respondents, process and chemical lists were used to prompt respondents. Two commenters [Ex. 3-625, Ex. 3-750] stated that the lists were incomplete and thereby biased the final data. However, one of these [Ex. 3-750] correctly stated that "...responders were also asked to volunteer additional processes or chemicals present in their plants." Since respondents did indeed volunteer "Other" chemicals, OSHA believes that the use of the prompt "Other" improved the final data instead of biasing it. The Inter-Industry Wood Dust Coordinating Committee commented that the survey did not include wood dust and processes specifically related to wood dust exposure, as prompts [Ex. 3-750]. However, respondents replied, in many instances, that there was exposure to "nuisance particulates". OSHA used these responses as surrogates for responses on wood dust. However, the Agency concluded that the costs in the Preliminary Regulatory Impact Analysis understated "the extent of new controls that would be needed in order to comply with the proposed wood dust standard" [Ex. 38A]. In SICs 24 and 25, the number of work stations where wood dust is found, the percent of work stations which would be out of compliance with the proposed levels, and unit costs for controlling exposure were revised to supplement the survey results, and the recalculated costs of compliance were provided to interested parties and entered into the docket [Ex. 38A].

Commenters also objected to the inclusion of non-production facilities in the survey [Ex. 3-1196, Ex. 3-877, Tr. 8/15/88, p. 105]. The sample survey was designed to represent the universe of facilities in each SIC. There are always a certain number of facilities in each SIC which are headquarters, distribution centers, or sales offices. Where workers at these facilities have no exposure to chemicals, there is no cost to control exposure and no benefits to accrue from lowered exposure levels. Inclusion of these facilities is statistically correct in order to represent that portion of the facilities in an SIC which would incur no cost.

The survey sample was statistically designed to include a higher proportion of larger establishments (20 or more employees) because of the wider variation in costs expected for large firms to comply. The American Mining Congress [Ex. 3-976] expressed concern about "the underrepresentation of small companies" in SICs 32 and 33, while the American Iron and Steel Institute [Ex. 3-1123] commented that average costs for large firms are not representative of costs for large steel facilities in SIC 33. OSHA believes that the generic nature of this rulemaking allows a greater latitude in grouping industries in order to estimate "average" costs, and that the higher proportion of large firms surveyed has provided a more valid estimate of the average costs. Small firms were not underrepresented. Rather, firms in the large size classes were "oversampled" using accepted statistical techniques.

Based on the survey, OSHA estimates that over 60 percent of production workers in most of the industries surveyed are potentially exposed to chemicals and about 10-15 percent of these would be overexposed at the levels proposed in this rulemaking. Chapter D presents OSHA's estimates of the benefits occurring from a reduction in the number of employees exposed to these chemicals.

The industry profiles that follow present economic information on industry sectors expected to be affected by the rulemaking. Most but not all of these industries were included in the sample survey. Table C-1, presented at the end of the chapter, contains employment and establishment data for each industry profiled. The number of establishments in that table was produced from 1985 Dun and Bradstreet data, to be consistent with the employment and economic impact data used in this chapter and in Chapter H.

Table C-4 shows the number of establishments estimated from the 1988 sample survey as compared to the number in the 1986-87 Dun and Bradstreet (D & B) file from which the sample was selected. In general, the estimated number of establishments from the survey is lower than the number in the original D & B file. Survey telephone contacts found that some sampled firms were either out of business, out of the scope of the survey (wrong SIC), or listed more than once on the file.


  Table C-4 - ESTABLISHMENT COUNTS FROM SAMPLE ESTIMATES AND ORIGINAL DUN
              & BRADSTREET

    NOTE: Because of its width, this Table C-4 has been divided; see
continuation for additional columns.
__________________________________________________________________________
                             ORIGINAL    NUMBER OF    NUMBER OF     SURVEY
                               NUMBER     ESTAB.'S     ESTAB.'S  ESTIMATED
                     SAMPLE  ESTAB.'S    FROM CNTY         FROM     NUMBER
ESTIMATION CELLS      UNITS  D&B FILE   BUS. PTRNS      BLS 202  OF ESTAB.
__________________________________________________________________________
CELL 1 SIC'S 243       160    13,486         6,510        8,333     12,150
CELL 2 SIC'S 245        74     1,512         1,101        1,114      1'099
CELL 3 SIC'S 249       115     5,364         3,682        3,638      4,868
CELL 4 SIC'S 25        154    16,129        10,812       10,041     12,804
CELL 5 SIC'S 26        309     8,228         6,324        6,731      7,022
CELL 6 SIC'S 27        234    78,345        56,137       57,928     60,282
CELL 7 SIC'S 281       155     2,797         1,342        1,445      2,013
CELL 8 SIC'S 282       123     1,562           648          845      1,380
CELL 9 SIC'S 283       176     2,288         1,277        1,407      1,740
CELL 10 SIC'S 284      137     4,202         2,375        2,359      3,656
CELL 11 SIC'S 285      104     1,789         1,433        1,419      1,693
CELL 12 SIC'S 286      124     1,399           918        1,028      1,224
CELL 13 SIC'S 287       71     1,752         1,028        1,377      1,421
CELL 14 SIC'S 289      110     3,733         2,724        2,367      3,299
CELL 15 SIC'S 291      122     1,055           442          796        722
CELL 16 SIC'S 295       88     1,160         1,294          917        960
CELL 17 SIC'S 299       73       735           581          461        628
CELL 18 SIC'S 307       80    13,876        12,112       11,762     12,316
CELL 19 SIC'S 301-306  158     3,187         1,857        1,884      2,778
CELL 20 SIC'S 3111      22       454           379          346        338
CELL 21 SIC'S 313-319   59     3,262         2,063        2,081      1,987
CELL 22 SIC'S 32       127    20,103        16,159       15,704     15,920
CELL 23 SIC'S 33       360     9,527         6,921        7,152      8,028
CELL 24 SIC'S 34       325    44,328        35,380       34,209     37,315
CELL 25 SIC'S 35,36,
  38,39                395   160,950        92,313        9,707    123,365
CELL 26 SIC'S 40,44,
  45,47                 82    52,277         -- --        -- --     42,025
CELL 27 SIC'S 46        63       624           605          953        577
CELL 28 SIC'S 49       368    18,430        17,725       16,019     15,812
CELL 29 SIC'S 5093,
  5153,5161,5191,5198  418    49,144        47,142       51,437     39,371
CELL 31 SIC'S 55,75     92   307,117       328,578      300,397    257,267
CELL 32 SIC'S 72,73    203   146,035         -- --        -- --    107,685
CELL 33 SIC'S 7641,
  7692                 107    18,399        10,162       13,514     15,095
CELL 34 SIC'S 80       402   261,380       390,223      377,887    222,843
CELL 99 SIC'S 37       102    14,958         9,498       11,438     13,617
__________________________________________________________________________

Table C-4  Establishment Counts from Sample Estimates and Original Dun &            Bradstreet - Continued
________________________________________________________
                       PERCENT      PERCENT      PERCENT
                        CHANGE       CHANGE       CHANGE
                    FROM ORIG.    FROM CNTY     FROM BLS
ESTIMATION CELLS        SAMPLE       SAMPLE       SAMPLE
________________________________________________________
CELL 1 SIC'S 243          -10%          87%          46%
CELL 2 SIC'S 245          -27%          -0%          -1%
CELL 3 SIC'S 249           -9%          32%          34%
CELL 4 SIC'S 25           -21%          18%          28%
CELL 5 SIC'S 26           -15%          11%           4%
CELL 6 SIC'S 27           -23%           7%           4%
CELL 7 SIC'S 281          -28%          50%          39%
CELL 8 SIC'S 282          -12%         113%          63%
CELL 9 SIC'S 283          -24%          36%          24%
CELL 10 SIC'S 284         -13%          54%          55%
CELL 11 SIC'S 285          -5%          18%          19%
CELL 12 SIC'S 286         -13%          33%          19%
CELL 13 SIC'S 287         -19%          38%           3%
CELL 14 SIC'S 289         -12%          21%          39%
CELL 15 SIC'S 291         -32%          63%          -9%
CELL 16 SIC'S 295         -17%         -26%           5%
CELL 17 SIC'S 299         -15%           8%          36%
CELL 18 SIC'S 307         -11%           2%           5%
CELL 19 SIC'S 301-306     -13%          50%          47%
CELL 20 SIC'S 3111        -26%         -11%          -2%
CELL 21 SIC'S 313-319     -39%          -4%          -5%
CELL 22 SIC'S 32          -21%          -1%           1%
CELL 23 SIC'S 33          -16%          16%          12%
CELL 24 SIC'S 34          -16%           5%           9%
CELL 25 SIC'S 35,36,
  38,39                   -23%          34%          33%
CELL 26 SIC'S 40,44,
  45,47                   -20%        -- --        -- --
CELL 27 SIC'S 46           -8%          -5%         -39%
CELL 28 SIC'S 49          -14%         -11%          -1%
CELL 29 SIC'S 5093,
  5153,5161,5191,5198     -20%         -16%         -23%
CELL 31 SIC'S 55,75       -16%         -22%         -14%
CELL 32 SIC'S 72,73       -26%        -- --        -- --
CELL 33 SIC'S 7641,
  7692                    -18%          49%          12%
CELL 34 SIC'S 80          -15%         -43%         -41%
CELL 99 SIC'S 37           -9%          43%          19%
________________________________________________________

It is possible to compare these survey establishment counts by comparing them to two alternative governmental sources: 1985 County Business Patterns from the U.S. Department of Commerce and 1987 ES-202 data from the Bureau of Labor Statistics (BLS), U.S. Department of Labor. Table C-4 also provides these governmental establishment counts.

In general, these two databases showed fewer establishments than either the survey or the D & B file. Some of these differences are due to the way an "establishment" is defined. D & B may split one establishment at a particular address into several establishments based on the various activities performed there; ES-202 and County Business Patterns may categorize the same establishment as one unit. Also, many state and federally-run establishments are included in the D & B file in the SIC related to their primary activity, rather than as governmental units, which would be the ES-202 and County Business Pattern classification.

There is no consensus among experts as to which source provides the most accurate establishment counts. Based on this comparison and other quality checks, OSHA believes that the survey has provided a sound basis for estimating the economic impact of this rulemaking.

SIC 20 - Food And Kindred Products

This major industry group includes establishments that manufacture or process food and beverages for human consumption as well as certain related products such as ice, chewing gum, vegetable and animal fats and oils, and prepared animal feeds [1, pp. 59 to 69]. This industry group was not included in the sample survey. Rather, industry data, costs and economic impact were estimated by experts familiar with this industry sector.

Employment and establishment data are shown in Table C-1. The total 1985 value of SIC 20 shipments ($301.6 billion) was 13 percent of the value of all manufacturing industry shipments; this represented the largest share of any two-digit manufacturing industry. The most important industry within SIC 20 is meat products, accounting for 22 percent of the value of shipments, followed by beverages and dairy products, accounting for 14 percent each [2, Vol 1:8].

In 1985, 1.6 million workers in over 29,000 establishments were employed in SIC 20. About 70 percent of these are production workers [Table C-1]. Employment has declined since 1979. The largest employer is the meat products industry, with 23 percent of the workforce in 1986, followed by preserved fruits and vegetables (15 percent) and beverages (13 percent). Meat and miscellaneous food products both experienced 1986 employment levels slightly above the 1979 peak [5]. The largest number of food products establishments are in the manufacturing or processing of miscellaneous foods and meat products (17 percent and 16 percent, respectively).

Establishments in SIC 20 are similar in size to those in the manufacturing industry as a whole, although there is a smaller concentration of very large establishments. Mean establishment size is 55 workers.

Most recent growth by larger food processors has been through business acquisitions rather than internal expansion. The food and beverage sector is becoming more concentrated and efficient. In most food industries for which data are available, concentration is moderate, with the largest four firms having a 30 percent share of sales. Exceptions can be found in cereal breakfast foods, where the four-firm concentration ratio is just over 75 percent, and in soft drinks, where it is 88 percent [3, pp. 33-1 to 39-39].

In the next few years, most food and beverage producers will benefit from increases in disposable income, favorable trends in consumer purchasing patterns, and continued low commodity prices. Decreased operating costs and expenses have resulted in a 6 percent increase in (revenue) income in 1985-86 for large food and beverage processors, despite sales gains of only a little more than 1 percent [3, p. 39-1].

In 1985, the median rate of return on assets in the food and kindred products industry was 5.1 percent; this was the third lowest for the 20 two-digit manufacturing industry group. The highest rates of return were registered by the cookie and cracker industry and the blended and prepared flour industry (11.8 percent and 11.1 percent, respectively), followed by the flavoring extracts industry (9.2 percent). At the other extreme, the wine and brandy industry registered a -0.9 percent rate of return on assets in 1985, with an average rate of under 0.1 percent for the 1984-86 period. The cheese and rice milling industries also have relatively low rates of return on assets (2.2 percent) [6].

OSHA received docket comments pertaining to several four-digit SICs (2011, 2013, 2016, 2017, and 2074) falling within the Food and Kindred Products industrial classification. Most comments addressed the use of three substances included in this rulemaking - carbon disulfide, ammonia, and chlorine - in the meat products sector (SIC 2013).

Commenters noted that firms in SIC 2013 produce hot dogs, luncheon meats, and boneless hams; production of these processed meats uses 3.6 billion pounds of meat and 0.9 billion pounds of poultry annually. Ten percent of all meat production goes into the production of processed meats; for some meats, the share is larger: 83 percent of all ham is processed into boneless hams [Exs. 3-421, 3-898]. Most of the meat used in processed meat is trimmings, which are not suitable for use in other meats. An estimated 65 percent of all processed meats are dependent on cellulosic materials for their manufacture [Ex. 3-421]. The production of casings of this type involves the use of carbon disulfide, and, according to commenters, achieving the proposed limit of 1 ppm for this substance would have created issues of technological and economic feasibility (discussed in greater detail in the Technological Feasibility chapter, below). In the final rule, the limits for carbon disulfide are 4 ppm as an 8-hour TWA and 12 ppm as a STEL; these limits should ameliorate any feasibility problems. In addition, the final rule has increased PEL and STEL limits for both ammonia (35 ppm STEL only) and chlorine (0.5 ppm TWA, 1.0 ppm STEL) which should further reduce the economic impact on this industry sector.

OSHA received many comments addressing the proposed 4-mg/m(3) TWA limit for grain dust (wheat, oats, and barley) in facilities classified in SIC 204, Grain Mill Products [Exs. 3-63, 3-110, 3-237, 3-299, 3-405, 3-752, and 3-755]. Comments were received from the owners of flour mills, rice mills, and feed mills. The National Feed and Grain Association (NFGA) [Ex. 3-752] estimated that the number of feed mills that use wheat, oats, or barley to produce feed is 1,260 facilities, or about 14 percent of all feed mills. The NFGA arrived at this estimate by assuming that feed mills use oats, wheat, and barley in proportion to the total U.S. usage of these grains as compared with the usage of other feed grains [Ex. 3-752]. (Estimates of the number of feed mills is difficult because feed mills are often classified in other industrial classifications if the major portion of their sales is generated by non-feed-mill activities, such as grain elevator operations or sales of farm supplies.) The NFGA characterized its estimate of 1,260 potentially affected feed mills as "conservative" [Ex. 3-752]. Commenters to the record from the Grain Mill Products segment of SIC 20 stated unanimously that achievement of the proposed 4 mg/m(3) limit for grain dust was not feasible in affected grain mills [Exs. 3-63, 3-110, 3-237, 3-299, 3-405, 3-752, and 3-755]. This issue is discussed further in the Technological Feasibility chapter. Because OSHA has revised the PEL for grain dust to 10 mg/m(3) in the final rule, most of the problems with technological feasibility raised by these commenters are likely to have been mitigated.

A comment to the record [Ex. 3-1080] from the National Cotton Council of America (NCCA) stated that cottonseed oil mills (SIC 2074) will be adversely impacted by the proposed rule. These mills process cottonseed and its various components for use in animal feed, edible oil, and cellulose products; their concerns are with the proposed limits for n-hexane and hexane insomers, vegetable oil mist, and grain dust. According to the NCCA, there are 50 operating cottonseed mills in the United States, and most of these are small, rural businesses without in-house industrial hygiene capability. The NCCA anticipates that its members will have difficulty measuring the proposed levels for these substances [Ex. 3-1080, p. 1]. The NCCA's comments are discussed in greater detail in Chapter F, Technological Feasibility.

SIC 21 - Tobacco Manufactures

Establishments in the tobacco manufactures industry produce cigarettes (SIC 211), cigars (SIC 212), chewing and smoking tobacco, and snuff (SIC 213), or they engage in tobacco stemming and redrying (SIC 214) [1, p. 70]. The major worker exposures in these industries are to particulates not otherwise regulated generated during the initial handling of tobacco and to chemicals that have been used to treat the tobacco. This industry was not included in the sample survey.

Data on employment and establishments for SIC 21 are shown in Table C-1. In 1985, the value of tobacco manufacturing shipments was $18.5 billion, slightly more than 6 percent of the value of shipments for all manufacturing [2, Vol. 1:8]. SIC 21 has less than 0.3 percent of the total employment or establishments in manufacturing [7, pp. 10, 15]. Three-quarters of the employees in this industry are production workers. The cigarette industry is the most important component of SIC 21, accounting for more than 80 percent of the value of shipments [2, Vol. 1: 8] and 70 percent of employment for this sector, but only 9 percent of establishments [7, p. 15].

Establishments in SIC 21 are large, with a mean size of 296 employees, compared to 55 for all manufacturing. More than half of the establishments in this two-digit SIC have fewer than 20 employees [Table C-1]; and more than 17 percent have 250 or more employees. The cigarette industry is especially highly concentrated, with a mean establishment size of 2,430 employees. Eleven establishments in the cigarette industry employ 1,000 or more workers, and 99.8 percent of all cigarette manufacturing employees work in these large establishments. Mean establishment sizes in other tobacco industries range from 80 to 135 employees [7, pp. 10, 15]. Employment in the tobacco products industry has declined every year since 1976 (except in 1981), with a total decline in employment of more than 23 percent over the last decade [5].

Most tobacco firms remain profitable because input costs have been relatively stable and prices have increased faster than consumption has declined. The major tobacco companies are continuing to reduce their vulnerability through mergers and diversification [3, pp. 40-1 to 40-7]. Thus, profitability in the tobacco manufactures industry is good. The 1985 median rate of return on assets (7.7 percent) was the fifth highest median rate of return on assets among firms in the 20 manufacturing industry groups [6].

OSHA received no comments or testimony on the tobacco manufacturing sector in the course of this rulemaking.

SIC 22 - Textile Mill Products

SIC 22 includes those establishments that perform any of the following six operations: 1) preparation of fiber and subsequent manufacturing of yarn, thread, braids, twine and cordage; 2) manufacturing broadwoven fabrics, narrow woven fabrics, knit fabrics, and carpets and rugs from yarn; 3) dyeing and finishing fiber, yarn, fabrics and knit apparel; 4) coating, waterproofing, or otherwise treating fabrics; 5) the integrated manufacturing of knit apparel and other finished articles from yarn; and 6) the manufacture of felt goods, lace goods, nonwoven fabrics, and miscellaneous textiles [1, p. 85].

According to the Department of Commerce, in 1986, shipments for the textile industry increased 4 percent. The value of shipments in 1985 ($53.3 billion) has increased 6 percent since 1981. Employment, however, remained on a long-term downward trend, although the 1986 drop was marginal. An upward trend in output and relatively high operating rates helped to keep the drop in employment to a minimum. Also, average hours worked, which increased in the second half of 1985, continued to rise in 1986 [3, p. 41-1].

Table C-1 presents data on the number of establishments and employment in SIC 22. Similar to other manufacturing industries, the mean establishment size in SIC 22 was 64 employees. Between 1981 and 1985, SIC 22 experienced a 15 percent decrease in employment. In 1985, almost 86 percent of the total number of employees were production workers [5]. The median rate of return on assets in the textile mill products industry was 5.6 percent in 1985 [6].

No commenters provided additional information on this industry, and it was not included in the 1988 sample survey.

SIC 23 - Apparel and Other Products

SIC 23 is referred to as the "cutting-up and needle trades," and includes establishments producing clothing and fabricating products by cutting and sewing purchased woven or knit textile fabrics and related materials. These materials may include leather, rubberized fabrics, plastics, and furs. In addition, establishments that manufacture clothing by cutting and joining materials are included [1, p. 97].

SIC 23 includes three types of apparel establishments: 1) the regular or inside factories, which perform the usual manufacturing functions within their own plant; 2) contract factories, which manufacture apparel from materials owned by others; and 3) apparel jobbers, which buy raw materials, design and prepare samples, arrange for the manufacture of clothing from their materials, and sell the finished product [1, p. 97]. According to U.S. Department of Commerce estimates, the 1987 value of shipments for SIC 23 experienced a growth rate of 5 percent over 1986 values [4, p. 45-1].

Between 1980 and 1985, SIC 23 was among the top ten SICs to experience the greatest employment decline. Due to large inventories at both retail and wholesale levels, and low consumer demand, there were decreases in both shipments and employment in 1985. In several geographic areas, plants were forced to close. The drop in employment has been attributed to the recent rise of imports into the U.S. market and to improvements in industry efficiency through streamlined operations and increased productivity [3, p. 42-2].

The apparel industry is a major employer of women and minorities, employing more than 6 percent of the manufacturing workforce in plants. Due to intense competition in the industry, profits and wages are lower in this industry than in most other manufacturing industries. The price of labor is the single most important cost component in the industry, which accounts for the sensitivity that employment levels have to industry growth levels. Production workers make up 85 percent of the apparel work force. Typically, as inventory levels grow, production slows down and employment drops [3, p. 42-2].

In 1986, current-dollar shipments in the apparel industry expanded in value by 3 percent. An increase in consumer demand was the major factor contributing to the upturn. Output levels began to regain former levels of output, and the falling rate of employment of about 1 percent was well below the 3.1 percent annual rate of decline during the 1980-1986 period [3, p. 42-2].

Table C-1 presents employment and establishment data for SIC 23 for 1985. During the period of 1981 through 1985, SIC 23 experienced a 10 percent decrease in employment. Almost 84 percent of the total number of employees were production workers [5]. In 1985, the median rate of return on assets in this SIC was 6.3 percent [6]. This sector was not included in the sample survey. Beginning with SIC 24 - Lumber and Wood Products, all the remaining major manufacturing SIC groups were included in the 1988 survey unless otherwise noted in the text.

SIC 24 - Lumber and Wood Products

This industry produces logs, pickets and fences, mining timbers, railroad ties, poles and pulpwood. SIC 24 includes establishments that cut timber and pulpwood, merchant sawmills, lath mills, shingle mills, cooperage stock mills, planing mills, and plywood mills and veneer mills engaged in producing lumber and wood basic materials; and establishments that manufacture finished articles made entirely or mainly of wood or related materials [1, p. 107]. According to U.S. Department of Commerce estimates, the logging industry's timber harvest in 1987 was an estimated $9.1 billion, compared with $8.8 billion in 1986. [4, p. 5-2].

The Department of Commerce reports that a strong expansion of the market for wood products took place in 1985 due to gains in housing and nonresidential construction activities. Although domestic demand for softwood lumber was strong, Canadian imports displaced American products and contributed to an oversupply, depressing prices. These lower prices lowered U.S. lumber producer profit margins and induced industrywide efforts to restrict imports of lower priced Canadian softwood lumber. In addition to the oversupply, accelerated harvesting to avoid pest damage forced inventories to go up and prices to fall further [3, p. 4-1].

In 1986, similar trends continued in the domestic market for wood products. This was due to a 6 percent rise in housing starts, continued growth in home remodeling and renovation, and strong demand from furniture markets and other end users. However, lower-priced softwood lumber imports from Canada continued to squeeze profits in 1986 [3, p. 4-1].

The Canadian softwood lumber prices brought about a trade agreement on December 30, 1986 between the United States and Canada, in which Canada agreed to set a 15 percent export tax on its softwood lumber. Canadian softwood lumber prices in the United States have risen 3 to 4 percent and imports have decreased about 3 to 4 percent. Since the agreement, Canada's market share has dropped from 33 percent to 28 percent. U.S. company earnings have increased despite a drop in housing starts. It is expected that the trade agreement will keep Canadian softwood prices up and continue to aid the domestic softwood lumber market [8].

Table C-1 presents employment and establishment data for SIC 24 for 1985, as well as for the three individual three-digit industry groups which were surveyed. In 1985, the mean establishment size in SIC 24 was 19 employees, significantly smaller than the average size in other manufacturing sectors. The median rate of return on assets in the SIC was 7.3 percent [6].

The National Kitchen Cabinet Association stated that the Dun and Bradstreet sampling frame used for the survey seriously underestimated the number of establishments [Ex. 80L]. Dun and Bradstreet estimated the number of establishments in SIC 24 as 36,710 [6]. The Department of Commerce in the 1985 County Business Patterns estimated the number to be 32,205 [7]. In making cost estimates for the Inter-Industry Wood Dust Coordinating Committee, National Economic Research Associates (NERA) assumed that there were only 26,485 establishments (using the 1982 Census of Manufacturers) [Ex. 3-748]. Thus, OSHA believes that the Dun and Bradstreet data used for the survey do not underestimate the total number of establishments.

Additionally, OSHA does not differ significantly from NERA on the total number of employees in SIC 24. OSHA used the Labstat Database of the U.S. Department of Labor to conclude that 697,000 persons were employed in 1985 [5]. NERA estimated that the industry employed 691,656 workers in 1986. Thus it appears that NERA accepts OSHA's estimate relating to the number of employees.

SIC 242 - Sawmills and Planing Mills

This SIC includes sawmills and planing mills, hardwood dimension and flooring mills, and special product sawmills. The U.S. Department of Commerce reported that in 1985, SIC 242 employed 26 percent of all employees and represented 21 percent of all establishments in SIC 24 [7]. The value of shipments from 1981 to 1985 rose 10 percent in SIC 2421 and 45 percent in SIC 2426 [4]. The Department of Commerce also reported that employment declined 14 percent in sawmills and rose 3 percent in hardwood dimension and flooring over the same time period. Production workers represent 87 percent of all employees in this industry. Special product sawmills (SIC 249) include facilities that produce shakes and shingles; approximately 290 firms use Western red cedar [9].

Firms in this sector were not included in the 1988 sample survey. OSHA relied on Dun and Bradstreet data to estimate the number of establishments in SIC 242. The National Dimension Manufacturers Association quoted the 1982 Census of Manufactures by stating that 789 establishments were in SIC 2426, of which 306 had 20 or more employees [Ex. 3-1160]. The Department of Commerce stated that there were 714 establishments in 1984, of which 320 had 20 or more employees [7]. Similarly the Census of Manufactures estimated the employment at 29,100 workers in 1987, while County Business Patterns estimated 26,841 in 1984. The difference in these estimates appears to be minor and largely associated with the difference in time and methods of data collection.

SIC 243 - Millwork, Veneer and Plywood

This SIC includes establishments that manufacture fabricated wood millwork, covered with materials such as metal and plastics. According to the U.S. Department of Commerce, the value of shipments for SIC 243 was $16.7 billion in 1985, which represents 31 percent of the value of shipments for SIC 24 [3]. The value of shipments in SIC 243 increased 27 percent since 1981. In 1985, the number of employees in SIC 243 was about 37 percent of SIC 24. The number of employees in SIC 243 increased by 18 percent from 1981 to 1985 [5]. Average hourly earnings dropped about 17 percent during that same time period. The number of establishments in SIC 243 in 1985 was about 38 percent of all establishments in SIC 24 [5].

In SIC 243, the 1988 survey identified more than twice as many small firms (fewer than 20 production workers) as large firms. In the small firms, maintenance work is performed for the most part by production workers. By contrast, in large firms, maintenance work is predominantly performed by dedicated maintenance workers.

The manufacturers classified in SIC 243 usually have one to three basic processes, with potential exposure to one to three substances. Thirty-nine percent of these processes involve exposure to chemicals or substances on an intermittent short-term basis (up to 30 minutes) with large firms tending to have more long-term exposures. Twenty-nine percent of the firms in this SIC reported the adoption of internal exposure standards. Of those small firms with internal exposure standards, most have adopted OSHA PELs. Nearly 71 percent of the large firms with standards reported using the OSHA PELs; the balance indicated that they rely on ACGIH TLVs or other standards. Employee monitoring had been performed at 17 percent of the processes.

The survey found that about 29 percent of the processes in SIC 243 are totally enclosed and 8 percent are located outdoors. Local exhaust ventilation is used most frequently to control exposures at processes not enclosed. Nearly 72 percent of the firms with chemical exposures have respirators for employee use, with a higher percentage of small firms reporting the presence of respirators than large firms. The combined data on exposure levels and methods of exposure control indicate that many plants which are estimated to incur some cost of compliance have overexposures in all processes at that plant.

Survey respondents in SIC 243 identified the presence of 35 different substances in SIC 243. Particulates not otherwise regulated were estimated to occur most frequently at a total of 8,956 processes. Particulates not otherwise regulated were identified in bleaching, coating/spraying/finishing/layup, cutting/sawing/planing, drying/baking, gluing/hot pressing, sanding/polishing/grinding, and metal working (rolling, milling, shaping). The final rule does not change the existing limit these particulates. Wood dust exposures occur in cutting/sawing/planing and sanding/polishing/grinding.

SIC 244 - Wood Containers

This SIC represents manufacturers of wood containers, including wood pallets and skids. The pallet industry is the third largest consumer of lumber in the United States, after the construction and furniture industries [Ex. 3-1125]. According to the U.S. Department of Commerce, more than 70 percent of the establishments in SIC 244 employ 20 or fewer people [7]. SIC 244 accounts for 6 percent of the establishments and employment in SIC 24. The total value of shipments in pallets and skids in 1987 was $1.5 billion, thus continuing the industry's third year of economic expansion [4, p. 5-9]. The number of establishments producing pallets and skids rose more than 67 percent from 1982 to 1986 [3, p. 4-10]. The National Wooden Pallet and Container Association (NWPCA) quoted the U.S. Forest Service's 1985 estimate of 2,340 firms in SIC 244 Ex. 3-899]. The 1984 County Business Patterns estimated 2,103 establishments [7]. The number of employees quoted by the NWPCA was 44,600, somewhat higher than Labstat's estimate of 40,500. However, County Business Patterns estimated 38,478 employees. Labstat estimated 40,500 employees in this industry in 1986 [5]. Hourly earnings of employees in pallets and skids rose 3 percent in 1986 to $6.32. This SIC was not included in the 1987 survey.

SIC 245 - Wood Buildings and Mobile Homes

This SIC includes manufacturers of wood buildings and mobile homes. The 1985 value of shipments for SIC 245 ($6.0 billion) represented 11 percent of the total value of shipments for SIC 24 [3]. The value of shipments in SIC 245 increased 6 percent from 1981 to 1985. In 1985, the number of employees in SIC 245 was 10.5 percent of SIC 24. Almost 77 percent of these employees are production workers. The number of establishments in SIC 245 in 1985 was 4.4 percent of all establishments in SIC 24 [5].

In SIC 245, the survey identified about as many small firms (fewer than 20 production workers) as large firms. In the small firms, maintenance work is generally performed by production workers. In large firms maintenance work is mostly performed by dedicated maintenance workers.

The manufacturers classified in SIC 245 usually have one to three basic processes, with potential exposure to one to three substances. Thirty percent of these processes involve exposure to these chemicals or substances on an intermittent short-term basis (up to 30 minutes) with large firms tending to have more long-term exposures. Almost thirty-seven percent of the firms in this SIC reported the adoption of internal exposure standards. Among small firms with internal exposure standards, most have adopted OSHA PELs. Nearly 67 percent of the large firms with standards reported using the OSHA PELs; the balance indicated that they rely on ACGIH TLVs. Employee monitoring had been performed at 40 percent of the processes.

The survey found that about 16 percent of the processes are totally enclosed and 16 percent are located outdoors. Local exhaust ventilation is used most frequently to control exposures at processes not enclosed. Nearly 78 percent of the firms with chemical exposures have respirators for employee use, with a higher percentage of large firms reporting the presence of respirators than small firms. The combined data on exposure levels and methods of exposure control indicate that many plants which are estimated to incur some cost of compliance have overexposures in all processes at that plant.

Survey respondents in SIC 245 identified the presence of 10 different substances. Particulates not otherwise regulated were estimated to occur the most frequently at a total of 703 processes. The final rule retains the existing limit for these particulates. Toluene, the second most frequently used chemical, was identified in coating/spraying/finishing/layup, gluing/hot pressing, and sanding/polishing/grinding. Wood dust occurs in cutting/sawing/planing and sanding/polishing/grinding operations.

SIC 249 - Miscellaneous Wood Products

This SIC covers miscellaneous wood products, and includes four four-digit SICs. SIC 249 represented 12 percent of the value of shipments for SIC 24 in 1985 [4]. The value of shipments in SIC 249 ($6.6 billion) increased almost 24 percent over 1981. In 1985, the number of employees in SIC 249 was about 15 percent of SIC 24. From 1981 to 1985, the number of employees in SIC 249 decreased by 4 percent. The number of establishments in SIC 249 in 1985 was about 14 percent of all establishments in SIC 24 [5].

SIC 2491 includes establishments that treat wood, sawed or planed in other establishments, with creosote or other preservatives to prevent decay and to protect against fire and insects. This industry also includes facilities that cut, treat, and sell poles, posts, and pilings. The Department of Commerce reports that during 1985 there was increased use of treated wood for home improvement projects, such as new decks and all-weather wood foundations. The market for railroad ties in 1985 was strong, as railroads replaced worn out ties. In 1986, however, sales of railroad ties declined. About 30 percent of total treated wood shipments are lumber and plywood [3, p. 4-14]. The Department of Commerce estimated that in 1986, the value of shipments in this industry increased by 7 percent [4, p. 5-15]. SIC 2491 represents 23 percent of the value of shipments for SIC 249. Employment rose in 1986 by 2.7 percent [3, p. 4-14]. The number of employees in SIC 2491 was almost 16 percent of SIC 249 and SIC 2491 represents almost 11 percent of all establishments in SIC 249 [6].

In SIC 249, the survey identified three times as many small firms (fewer than 20 production workers) as large firms. In the small firms, maintenance work is performed for the most part by production workers or a dedicated maintenance staff. Large firms primarily use dedicated maintenance workers to perform maintenance duties.

The manufacturers classified in SIC 249 usually have one to four basic processes, with potential exposure to one to four substances. Thirty-four percent of the processes in this SIC involve exposure to these chemicals or substances on an intermittent short-term basis (up to 30 minutes), with large firms tending to have more long-term exposures. Forty-three percent of the firms in this SIC reported the adoption of internal exposure standards. Of these, most small firms have adopted OSHA PELs. Nearly 83 percent of the large firms with standards reported using the OSHA PELs; the balance indicated that they rely on ACGIH TLVs. Employee monitoring had been performed at 11 percent of the processes.

The survey found that about 22 percent of the processes are totally enclosed and 11 percent are located outdoors. Local exhaust ventilation is used most frequently to control exposures at processes not enclosed. Roughly 48 percent of the firms with chemical exposures have respirators for employee use, with a higher percentage of large firms reporting the presence of respirators than small firms. The combined data on exposure levels and methods of exposure control indicate that many plants which are estimated to incur some cost of compliance have overexposures in all processes at that plant.

Survey respondents identified the presence of 25 different substances in SIC 249. Particulates not otherwise regulated were estimated to occur most frequently at a total of 1,678 processes. Particulates were identified in coating/spraying/finishing/layup, cutting/sawing/planing, drying/baking, gluing/hot pressing, sanding/polishing/grinding, stamping/shaping/molding/pressing, and assembly. Wood dust exposures occur in cutting/sawing/planing and sanding/polishing/grinding.

SIC 25 - Furniture and Fixtures

Manufacturers of household, office, public building, and restaurant furniture and office and store fixtures are included in SIC 25 [1, p. 114]. The U.S. Department of Commerce states that producers of furniture and fixtures recently have benefited from lower real interest rates, a reduction in the value of the dollar versus other major currencies, and changes in the tax laws [3]. In addition, the U.S. furniture industry is undergoing consolidation; big firms are becoming larger and accounting for a greater share of the market. The remaining smaller firms are finding it more difficult to compete, given the rapid increase in low-priced imports. Moreover, new manufacturing technologies require large capital investments and large volume, neither of which are readily available to small firms [3, p. 44-2].

For the industry, the value of shipments in 1985 increased by 31 percent over the level in 1981. In household furniture, the value of shipments for 1987 increased an estimated 7 percent following a growth of 5.4 percent in 1986 [4, p. 47-2]. Although furniture manufacturers anticipate stronger demand in the future, these manufacturers remain uncertain as to the duration and extent of increased demand. Therefore, rather than hiring additional workers, producers have increased the average number of hours worked by current employees. This trend was evident in the wood and metal furniture plants, where average overtime hours increased 16 percent and 24 percent, respectively, in the first half of 1986 [3, p. 44-2].

Table C-1 presents employment and establishment data for SIC 25 for 1985. Almost 80 percent of the total number of employees working in SIC 25 were production workers and the median rate of return on assets in the furniture industry was 7.3 percent in 1985.

In SIC 25, the survey detected twice as many small firms (fewer than 20 production workers) as large firms. In small firms, maintenance work is performed for the most part by production workers, whereas large firms primarily use dedicated maintenance workers.

The manufacturers classified in SIC 25 usually have one to four basic processes, with potential exposure to as many as six substances. Twenty-four percent of these processes involve exposure to these chemicals or substances on an intermittent short term basis (up to 30 minutes), with large firms tending to have more long-term exposures. Fifty-four percent of the firms in this SIC reported the adoption of internal exposure standards. Among small firms with internal exposure standards, most use the OSHA PELs. About 66 percent of the large firms with standards reported using the OSHA PELs; the remainder indicated that they rely on ACGIH TLVs. Employee monitoring had been performed at 19 percent of the processes.

The survey found that about 25 percent of the processes are totally enclosed and 3 percent are located outdoors. Local exhaust ventilation is used most frequently to control exposures at processes not enclosed. Nearly 78 percent of the firms with exposures provide respirators for employee use, with a higher percentage of small firms reporting the presence of respirators than large firms. The combined data on exposure levels and methods of exposure control indicate that many plants which are estimated to incur some cost of compliance do not have overexposures in all processes at that plant.

Survey respondents identified the presence of 36 different substances in SIC 25. Particulates not otherwise regulated were estimated to occur most frequently at a total of 3,433 processes. Toluene, the second most frequently used chemical, was identified at processes in coating/spraying/finishing/layup, drying/baking, gluing/hot pressing, drilling/boring and sanding/polishing/grinding. Wood dust exposures occur in cutting/sawing/planing, drilling/boring, and sanding/polishing/grinding.

SIC 26 - Paper and Allied Products

Establishments in this industry process fiber from trees, wastepaper, and other fibrous materials into end products that are used by both consumers and industry [1, p. 100]. Based on U.S. Department of Commerce estimates, the paper and allied products industry experienced an increase of 16 percent in the value of shipments from 1981 to 1985, and over 10 percent between 1985 and 1986 [3]. Net profits for 27 paper industry firms were reported to have averaged nearly 60 percent higher in the first six months of 1987 than in the first half of 1986 [4, p.6-1].

The industry's overall demand patterns are closely linked to rates of change in GNP. In 1985, for example, real growth for the industry was judged to be flat, trailing that of the GNP. The largest fluctuations in the industry's shipments have occurred in products geared specifically for commercial-industrial use, which are tied to the annual rate of business activity [3, p. 5-1].

Table C-1 presents employment and establishment data for SIC 26 for 1985. From 1981 to 1985, employment declined by approximately 2 percent. Almost 76 percent of the total number of employees were production workers [5]. In 1985, the median rate of return on assets was 7.4 percent [6].

Within SIC 26, there are six, three-digit SIC groups. SIC 261 includes manufacturers of pulp from wood or other materials. The Department of Commerce reports that U.S. market pulp prices dropped nearly 10 percent in the first six months of 1985. By the end of 1985, however, producers' pulp mill inventories had dropped, helping to stabilize pulp prices. About one-fourth of all market pulp companies either shut down some of their mills in 1985 or curtailed production to reduce the oversupply in the market. In 1986, the industry experienced increased productivity, higher prices and improved worldwide demand. For SIC 261, the value of shipments in 1987 increased by 2.7 percent over 1986. SIC 261 represents 3.5 percent of the value of shipments for SIC 26 [3, p.5-2].

SIC 262 includes manufacturers of paper from wood pulp and other fiber pulp, and manufacturers of converted paper products. SIC 263 includes manufacturers of paperboard. SIC 262 represents 11 percent of the value of shipments for SIC 26. The value of shipments decreased by 3.6 percent. The number of employees in SIC 263 was less than 1 percent of SIC 26 [5].

SIC 264 includes manufacturers of coated or laminated flexible materials used for packaging purposes. In this sector, the value of shipments, which represents 36 percent of the value of shipments for SIC 26, increased by 17 percent during the same period. The number of employees in SIC 264 was 34 percent of SIC 26 [5]. SIC 265 includes manufacturers of setup paperboard boxes from purchased paperboard. Corrugated boxes have taken the place of wooden shipping containers, pallets, and metal drums in the U.S. packaging market in recent years [3, p. 5-6]. Similarly, consumption of folding boxes continued steadily in 1985. This pattern continued in 1986 with shipments of corrugated boxes increasing 5.5 percent and 3 percent for folding boxes. Several important nondurable end users of folding cartons, such as producers of beverages, dry foods, textiles, sporting goods and toys, hardware, candy, and cosmetics, showed significant declines in real growth in 1985, while the market for boxed paper goods either grew slightly or remained fairly level [3, p. 5-9].

Manufacturers of sanitary food containers, such as paperboard milk cartons and paper serving and eating utensils, are also included in SIC 265. This industry has been strongly influenced by the shift to plastic containers. Having experienced two successive years of decline, the industry increased the value of shipments by 2 percent in 1986. Since 1983, the most rapid growth area within the sanitary food container industry has been aseptic packaging. This is specially treated paperboard combined with plastic film and aluminum foil.

The value of shipments for SIC 265 increased by 16 percent from 1981 to 1985. This three-digit SIC represents 24 percent of the value of shipments for all of SIC 26. In 1985, the number of employees in SIC 265 was 29 percent of SIC 26 [5]. SIC 266 includes manufacturers of building paper and building board from wood pulp and other fibrous materials. Trends in employment and value of shipments have followed overall trends in SIC 26.

In SIC 26, the survey identified half as many small firms (fewer than 20 production workers) as large firms. In small firms, maintenance work is performed largely by production workers, although some firms use outside contractors. Large firms generally use a separate maintenance staff to perform maintenance duties.

The manufacturers classified in SIC 26 usually have one to six basic processes, with potential exposure to as many as seven chemicals or substances. Twenty-nine percent of these processes involve exposure to these chemicals or substances on an intermittent short-term basis (up to 30 minutes), with large firms tending to have more long-term exposures. firms in this SIC are equally divided between those adopting no internal exposure standards and those adopting OSHA PELs. Among small firms with internal exposure standards, all have adopted OSHA PELs. Nearly 81 percent of the large firms with standards reported adopting the OSHA PELs; the balance indicated that they rely on ACGIH TLVs. Employee monitoring had been performed at 36 percent of the processes. The survey found that about 25 percent of the processes are totally enclosed and 4 percent are located outdoors. Local exhaust ventilation is used most frequently to control exposures at processes not enclosed. Approximately 42 percent of the firms with chemical exposures have respirators for employee use, with a higher percentage of small firms reporting the presence of respirators than large firms. The combined data on exposure levels and methods of exposure control indicate that many plants which are estimated to incur some cost of compliance have overexposures in all processes at that plant.

Survey respondents identified the presence of 46 different chemicals or substances in SIC 26. Particulate not otherwise classified occurred the most frequently at a total of 1,753 processes. Cellulose was identified at 664 processes. The final rule retains the existing limits for both particulates not otherwise regulated and cellulose. Wood dust exposures occurred in cutting/sawing/planing.

SIC 27 - Printing, Publishing, and Allied Industries

This industry is divided into a publishing sector, a printing sector and a sector of related services. The publishing sector includes newspaper publishing (SIC 271), periodical publishing (SIC 272), book publishing (SIC 2731), miscellaneous publishing (SIC 274) and greeting card publishing (SIC 277). The printing sector includes commercial printing (SIC 275), book printing (SIC 2732), and printing trade services (SIC 279). The related services sector includes manifold business forms (SIC 276) and blankbooks and bookbinding (SIC 278) [1, pp. 106-110].

There were approximately 84,279 establishments in the printing and publishing business in 1985. The majority of these firms (84.1 percent) had fewer than 20 employees, and the mean establishment size was 17 employees. The firms in SIC 27 had 1.4 million employees and 789,000 production workers [Table C-1]. According to the U.S. Department of Commerce, the value of shipments for all printing and publishing establishments in 1986 ($118.6 billion) was 5.2 percent of the value of shipments for all manufacturing industries. Most of the value of shipments in SIC 27 is from the commercial printing sector (32.1 percent) [4]. In 1985, the median rate of return on assets was 8.2 percent for the printing and publishing industry [6].

Foreign trade has not been a major concern for this industry in the past, but imports are beginning to increase at a steady rate. The respective values of imports and exports were very close in 1987, with $1.6 billion in imports and $1.5 billion in exports [3, p. 29-2].

The newspaper industry has improved its performance after several years of slow growth. The value of shipments for SIC 271 was $29.2 billion in 1986. Total employment rose an estimated 2.2 percent in 1986 to 420,000 employees, but production employment remained virtually unchanged at 151,900 employees. Sales revenues increased by 8.9 percent, from $14.8 billion in 1986 to $16.2 billion in 1987. Advertising revenues rose slightly, but most of this gain was due to rate increases and growth in classified ad volume. Total net worth increased by 14.2 percent from 1986 to 1987 [10].

The periodical industry has experienced moderate growth in both advertising receipts and circulation. Advertising revenue increased about 4 percent in 1987, while circulation revenues increased slightly due to the increase in subscriptions for consumer magazines. There was another large increase in the number of new publications entering the market; over 250 new periodicals were published in 1987 [3, p. 29-6]. The value of shipments of the periodical industry was $15.7 billion in 1986, an increase of 3.1 percent over the 1985 figure of $15.2 billion. The total number of employees in the periodical industry increased in 1986 (98,100 employees), while the number of production workers decreased (14,200 employees). The periodical industry has the lowest ratio of production workers to total employees (14.5 percent) within SIC 27.

The commercial printing industry (SIC 275) has been very profitable over the last decade. The 1987 value of shipments ($40.9 billion) increased 7.5 percent over the 1986 value of shipments ($38.0 billion). Between 1980 and 1985, the value of shipments increased by 11.5 percent compounded annually. Total employment and production employment have also been increasing substantially from 1986 to 1987 (3.7 percent and 2.7 percent, respectively). The outlook for this industry is steady growth [3, pp. 29-12 to 29-14].

Both book publishing and printing showed strong gains over the last several years. Value of shipments and total employment increased by 5.5 percent and 0.7 percent, respectively, from 1985 to 1986. Spurred by the increase in school enrollment, sales of textbooks were projected to reach 29 percent of total industry sales in 1988. Book printing usually follows the path of book publishing, increasing substantially when book publishing has a strong year [3, pp. 29-9 to 29-13].

Miscellaneous publishing and printing consists of newsletters, catalogs, directories, greeting cards, and business forms. This industry has seen steady gains due in part to the success of mail-order catalogs, telephone directories, and newsletters [3, pp. 29-13 to 29-19].

In this SIC, the survey identified six times as many small firms (fewer than 20 production workers) as large firms. In the small firms, approximately two-thirds of maintenance work is performed by production workers. Outside contractors do approximately one-fourth of maintenance work, and maintenance staff and other sources make up the remainder. Large firms divide maintenance work about equally between a dedicated maintenance staff, production workers, and outside contractors.

The manufacturers classified in this SIC usually have one to three basic processes, with potential exposure to as many as six chemicals. Employees are exposed to these chemicals on an intermittent short term basis (up to 30 minutes) or continuously (up to 8 hours per day) with large firms tending to have more long-term exposures. Small firms generally have no internal exposure standards; when they do, the OSHA PELs are followed about fifteen percent of the time. Over one-half of large firms reported using the OSHA PELs; the balance indicated that they have no standards or they rely on ACGIH TLVs or NIOSH RELs. Air monitoring data were provided for about one-tenth of the processes found in all plants, and for about one-third of the processes found in large firms.

The survey found that about two-thirds of the processes are totally enclosed and less than one percent are located outdoors. Local exhaust ventilation and general dilution are used most frequently to control exposures at processes not enclosed or outdoors. In less than five percent of the firms with chemical exposures, production workers use respirators with a higher percentage of large firms using respirators than small firms.

Survey respondents identified isopropyl alcohol, stoddard solvent, and methyl alcohol among the chemicals most prevalent in this SIC. These are used in lithographic printing and platemaking and letterpress printing which were the processes most frequently listed by respondents. Toluene, xylene, and trichloroethylene were also identified in the survey. A large commercial printer, R.R. Donnelly and Sons, confirmed the presence of toluene in press operations and expressed concern over the ability to meet the proposed levels, especially during cleaning [Ex. 3-916].

SIC 28 - Chemicals and Allied Products

SIC 28 includes establishments that produce basic chemicals, and establishments that manufacture products using chemical processes. There are three general classes of products: 1) basic chemicals, such as acids, alkalies, salts, and organic chemicals; 2) chemical products to be used in further manufacturing, such as synthetic fibers, plastics materials, dry colors, and pigments; and 3) finished chemical products to be used for consumption, such as drugs, cosmetics, and soaps; or to be used as materials or supplies in other industries, such as paints, fertilizers, and explosives [1, p. 132].

The chemical and allied products industries have experienced small but steady growth over the recent past. Total shipments by the chemical industry increased approximately 3.1 percent in 1987, following a 3.5 percent gain in 1986 [3]. Chemical prices have been stable since 1982, due to steady or declining energy costs. Like many other U.S. industries, various sectors within the chemical industry are undergoing structural changes, such as mergers, plant closings, sale of plants, and other adjustments. This industry employs approximately 5 percent of all industry workers, but more than 10 percent of all U.S. scientists and engineers. SIC 28 experienced a 6 percent decline in employment between 1981 and 1985. In 1985, 55.4 percent of the total number of employees in SIC 28 were production workers. The value of shipments increased 8.9 percent during the 1981 to 1985 time period. The median rate of return on assets in the chemical industry was 6.3 percent [6].

Within SIC 28, there are eight, three-digit SICs, which are described below.

SIC 281 - Industrial Inorganic Chemicals

This SIC includes establishments that manufacture basic industrial inorganic chemicals. SIC 281 represented 10.3 percent of the value of shipments of SIC 28 in 1985 [2]. The value of shipments increased 12.9 percent since 1981, and employment declined by 12 percent. Production workers equaled almost 51 percent of all workers. The number of establishments in SIC 281 was 14.5 percent of all establishments in SIC 28 [Table C-1].

SIC 281 is subdivided into four groups. Examples of the products of each four-digit SIC are given below. SIC 2812 Products - Chlorine, soda ash, caustic potash, caustic soda, washing soda, and sodium bicarbonate. SIC 2813 Products - Oxygen, acetylene, argon, carbon dioxide, and hydrogen. SIC 2816 Products - Color pigments, iron colors, iron oxide, lead oxide pigments, mineral colors, titanium pigments, and zinc oxide pigments. SIC 2819 Products - Sulfuric, hydrochloric, and hydrofluoric acids.

In SIC 281, the survey identified three times as many small firms (fewer than 20 production workers) as large firms. In the small firms, maintenance work is performed for the most part by production workers, although some firms employ dedicated maintenance workers. Large firms predominantly employ workers specifically for maintenance duties.

The manufacturers classified in this SIC usually have one to two basic processes, with potential exposure to as many as six substances. Fifty-two percent of these processes involve exposure to chemicals or substances on an intermittent short-term basis (up to 30 minutes), with large firms tending to have longer-term exposures. Most firms in SIC 281 reported the adoption of OSHA PELs as their internal standards. Employee monitoring had been performed at 67 percent of the processes. The survey found that about 32 percent of the processes are located outdoors. Local exhaust ventilation is used most frequently to control exposures at processes not enclosed. About 24 percent of the firms with chemical exposures have respirators for employee use, with a higher percentage of small firms reporting the presence of respirators than large firms. The combined data on exposure levels and methods of exposure control indicate that most plants which are estimated to incur some cost of compliance do not have overexposures in all processes at that plant.

Survey respondents identified the presence of 58 different substances in SIC 281. Carbon dioxide was estimated to occur the most frequently at a total of 292 processes. Carbon dioxide was identified in recovery/reprocessing, packaging/bagging, loading/offloading/receiving/ handling, process inspection, reaction/fermentation, and separation. Another common substance, sodium hydroxide, was identified in boilers.

SIC - Plastics Materials and Synthetics

This SIC includes manufacturers of plastics materials and synthetic resins, synthetic rubbers, and cellulosic and other manmade fibers. Plastics make up a variety of products which are used in diverse markets. Packaging and construction account for over 50 percent of consumption, with the remainder going into the transportation, electronics, and medical industries [3]. SIC 282 represents almost 17 percent of the value of shipments of SIC 28. The value of shipments in SIC 282 increased 9.2 percent over the period 1981 to 1985 [2]. Industry shipments of plastics in 1986 gained 6.3 percent as volume rose in response to slightly increased demand for materials. However, declining prices of plastic materials held shipments to a 2 percent increase [3, p. 14-1].

Table C-1 gives employment and establishment data for this segment. The number of employees in SIC 282 in 1985 was almost 16 percent of SIC 28 and the number of establishments was 8 percent of all establishments in that SIC. In 1985, employment in SIC 282 declined by 12 percent, and production workers equaled 66.5 percent of all workers [5].

SIC 282 is subdivided into four groups. Examples of the products from each of these four-digit SICs are given below. SIC 2821 Products - Cellulose plastics materials, phenolic and other tar acid resins, acrylic resins, polyethylene resins, coumarone-indene and petroleum polymer resins, and casein plastics.

SIC 2822 Products - Copolymers of butadiene and styrene, or butadiene

and acrylonitrile, and polybutadienes. SIC 2823 Products - Cellulose, rayon, and triacetate fibers. SIC 2824 Products - Fibers of acrylic, acrylonitrile, polyvinyl ester, and nylon. In SIC 282, the survey identified twice as many small firms (fewer than 20 production workers) as large firms. In small firms, maintenance work is either performed by production workers or dedicated maintenance workers. Large firms primarily employ workers specifically for maintenance duties.

The manufacturers classified in SIC 282 usually have one to six basic processes, with numerous firms having exposures to as many as six different substances. Forty-five percent of these processes involve exposure to these chemicals or substances on an intermittent short-term basis (up to 30 minutes), with large firms tending to have more short-term exposures. Most firms in this SIC have adopted OSHA PELs as their internal standards. Of the small firms with internal exposure standards, most have adopted OSHA PELs or ACGIH TLVs. About 49 percent of large firms reported using OSHA PELs, with 36 percent reporting the adoption of ACGIH TLVs.

The survey found that about 33 percent of the processes are totally enclosed and 24 percent are located outdoors. Local exhaust ventilation is used most frequently to control exposures at processes not enclosed. More than 28 percent of the firms with chemical exposures have respirators, with an equal percentage of small and large firms reporting the availability of respirators for employee use. The combined data on exposure levels and methods of exposure control indicate that many plants which are estimated to have some cost of compliance have overexposures in some, but not all, processes in the plant.

Survey respondents identified the presence of 53 different substances in SIC 282. Styrene was estimated to occur the most frequently at a total of 209 processes. Styrene was identified in recovery/reprocessing/reclamation, drying/baking, separation, blending/mixing/formulating, packaging/bagging, extrusion, crushing/grinding/calcining, loading/offloading/receiving/handling, and reaction/fermentation. Another common substance in SIC 282 was isopropyl alcohol, which occurred in eight different processes.

SIC - Drugs

This group includes establishments that manufacture, fabricate, or process medicinal chemicals and pharmaceutical products. The value of shipments in SIC 283 has increased 40 percent from 1981 to 1985 [2]. SIC 283 represents 16 percent of the value of shipments of SIC 28 and almost 20 percent of the number of employees. The U.S. Department of Commerce estimated that the pharmaceutical industry experienced a 6.3 percent increase in the value of shipments in 1986. However, after adjusting for price changes, this growth rate was closer to 1.8 percent. Productivity also increased in 1986, growing by approximately 2.6 percent [3, p. 17-1].

As seen in Table C-1, the number of establishments in SIC 283 was almost 12 percent of all establishments in SIC 28. Employment increased by 3 percent since 1981, and production workers equaled approximately 46 percent of all workers in SIC 283. Agar, vitamins, antibiotics, vaccines, and viruses are examples of the products of this SIC.

In SIC 283, the survey identified three times as many small firms (fewer than 20 production workers) as large firms. In the small firms, maintenance work is generally performed by either dedicated maintenance workers or by general production workers. In large firms, most maintenance work is performed by workers specifically employed for maintenance duties.

The manufacturers classified in SIC 283 usually have one to five basic processes, with potential exposure to one to three substances. Fifty percent of all employees are exposed to these chemicals or substances on an intermittent basis (up to 30 minutes), with small firms tending to have more long-term exposures. Among small firms with exposure standards, most have adopted OSHA PELs. Among large firms, a significant percentage have adopted ACGIH TLVs, although most still rely on OSHA PELs. Employee monitoring had been performed at 34 percent of the processes.

The survey found that about 53 percent of the processes are totally enclosed and 4 percent are located outdoors. Local exhaust ventilation is used most frequently to control exposures at processes not enclosed. In 21 percent of the firms with chemical exposures, respirators were available for employee use, with a higher percentage of large firms reporting the presence of respirators than small firms. The combined data on exposure levels and methods of exposure control indicate that very few plants which are estimated to incur some cost of compliance have overexposures at all processes in that plant.

Survey respondents identified the presence of 40 different substances in SIC 283. Isopropyl alcohol was estimated to occur the most frequently at a total of 577 processes. Isopropyl alcohol was identified in boilers, coating/spraying/finishing/layup, drying/baking, blending/mixing/formulating, packaging/bagging, loading/offloading/receiving/handling, reaction/fermentation, and separation.

SIC 284 - Soaps, Cleaners, and Toilet Goods

This SIC includes manufacturers of detergents, emulsifiers, cosmetics, and producers of glycerin. SIC 284 represents 15 percent of the value of shipments of SIC 28 [3]. The value of shipments in SIC 284 increased 20 percent from 1981 to 1985 [2]. In 1986 the value of shipments was estimated at $31 billion, which represents about a 4 percent increase over 1985 values [3, p. 16-1].

The number of employees in this SIC was almost 15 percent of SIC 28 and the number of establishments was almost 22 percent. In 1985, employment in SIC 284 had increased by 1 percent since 1981, and production workers equaled approximately 63 percent of all workers in SIC 284 [5].

There are four subgroups within SIC 284. Examples of the products produced by each four-digit SIC are given below. SIC 2841 Products - Soap, synthetic organic detergents, inorganic alkaline detergents, and crude and refined glycerin from vegetable and animal fats and oils.

SIC 2842 Products - Household, institutional, and industrial plant

disinfectants, non-personal deodorants, dry cleaning preparations, household bleaches, and other sanitation products.

SIC 2843 Products - Textile and leather finishing agents, soluble oil and

greases. SIC 2844 Products - Perfumes, cosmetics, home permanent kits, shampoos, shaving products, and talcum powder. In SIC 284, the survey identified twice as many small firms (fewer than 20 production workers) as large firms. Maintenance work in small firms is basically performed by production workers, while dedicated maintenance workers, and in some firms production workers, handle this task in large firms.

The manufacturers classified in SIC 284 usually have one to three basic processes, with potential exposure to one to eight substances. Fifty percent of all employees are exposed to these chemicals or substances on an intermittent basis (up to 30 minutes), with large firms tending to have more long-term exposures. Most firms in this SIC have adopted OSHA PELs. Employee monitoring had been performed at 26 percent of the processes.

The survey found that roughly 38 percent of the processes are totally enclosed and 9 percent are located outdoors. Local exhaust ventilation is used most frequently to control exposures at processes not enclosed. Nearly 44 percent of the firms with chemical exposures have respirators for employee use, with a higher percentage of large firms reporting the presence of respirators than small firms. The combined data on exposure levels and methods of control indicate that very few plants which are estimated to incur some cost of compliance have overexposures at all processes at that plant.

Survey respondents identified the presence of 52 different substances in SIC 284. Sodium hydroxide was estimated to occur most frequently at a total of 452 processes. Sodium hydroxide was identified in drying/baking, blending/mixing/formulating, packaging/bagging, loading/offloading/receiving/handling, reaction/fermentation, and separation.

SIC 285 - Paints and Allied Products

This SIC includes manufacturers of paints and allied paint products such as varnishes, shellacs, and paint removers. The paint industry grew by about 5.3 percent in 1986, compared to 1985's decline of 2.9 percent [3, p. 15-1]. Estimated shipments for 1986 were $11.1 billion, of which architectural coatings accounted for about 41 percent, followed by product coatings (35 percent) and specialty products (24 percent) [3, p. 15-2]. SIC 285 represents about 6 percent of the value of shipments of SIC 28. The value of shipments increased almost 26 percent from 1981 to 1985 [2]. The number of employees in SIC 285 was 6 percent of SIC 28 and the number of establishments was 9 percent.

In SIC 285, the survey identified three-fourths as many small firms (fewer than 20 production workers) as large firms. In small firms, maintenance work is either performed by production workers or dedicated maintenance workers. Large firms predominantly use workers dedicated to maintenance duties.

The manufacturers classified in SIC 285 usually have one to five basic processes, with numerous firms having potential exposures to as many as seven different substances. About 40 percent of the employees are exposed to these chemicals or substances on an intermittent short-term basis (up to 30 minutes), with large firms tending to have longer-term exposures. Most firms in this SIC have adopted OSHA PELs or ACGIH TLVs as their internal standard; about 58 percent of the firms reported using OSHA PELs and 19 percent reported the adoption of ACGIH TLVs.

The survey found that about 34 percent of the processes are totally enclosed and 11 percent are located outdoors. Local exhaust ventilation is used most frequently to control exposures at processes not enclosed. About 17 percent of the firms with chemical exposures have respirators for employee use, with a higher percentage of large firms reporting the presence of respirators. The combined data on exposure levels and methods of exposure control indicate that most plants which are estimated to incur some cost of compliance have overexposures in some, but not all, processes at that plant.

Survey respondents identified the presence of 39 different substances in SIC 285. Stoddard solvent was estimated to occur most frequently at a total of 941 processes. Stoddard solvent was identified in recovery/reprocessing/reclamation, coating/spraying/finishing/layup, drying/baking, blending/mixing/formulating, packaging/bagging, crushing/grinding/calcining, loading/offloading/receiving/handling, reaction/fermentation, and separation. Another common substance in this SIC was ethylene glycol, which occurred in seven processes.

SIC 286 - Industrial Organic Chemicals

This SIC includes manufacturers of a variety of industrial organic chemicals. Industry shipments of organic chemicals increased approximately 2 percent over 1985, which was the same level of growth experienced in the previous year [3]. In 1985, the value of shipments for SIC 286 was $41.8 million, representing 21 percent of the value of shipments of SIC 28 [2]. The number of employees in SIC 286 was almost 11 percent of SIC 28 and the number of establishments was approximately 7 percent. Employment in SIC 286 increased by 10 percent, and production workers equaled 51 percent of all workers [.Table C-1].

There are three subgroups in SIC 286. Examples of products for each four-digit SIC are given below. SIC 2861 Products - Hardwood and softwood distillation products, wood and gum naval stores, charcoal, natural dyestuffs and natural tanning materials.

SIC 2865 Products -- Toluene, benzene, synthetic organic dyes and

pigments. SIC 2869 Products -- Alcohols, caprolactam, and ethylene glycol.

In SIC 286, the survey identified three-fourths as many small firms (fewer than 20 production workers) as large firms. Small firms primarily use production workers to perform maintenance tasks. Large firms, on the other hand, primarily use dedicated maintenance workers to perform maintenance duties. Some small and large firms use outside contractors.

The manufacturers classified in this SIC usually have two to four basic processes, with potential exposure to as many as six substances. Fifty-six percent of the employees are exposed to these chemicals or substances on an intermittent short-term basis (up to 30 minutes), with large firms tending to have longer-term exposures. Most firms in SIC 286 have adopted OSHA PELs or ACGIH TLVs as their internal standards. Employee monitoring had been performed at 78 percent of the processes.

The survey found that about 34 percent of the processes are totally enclosed and that nearly 38 percent of the processes are located outdoors. Local exhaust ventilation is used most frequently to control exposures at processes not enclosed. Roughly 34 percent of the firms with chemical exposures have respirators for employee use, with a higher percentage of large firms reporting the presence of respirators than small firms. The combined data on exposure levels and methods of exposure control indicate that very few plants which are estimated to incur some cost of compliance have overexposures in all processes at that plant.

Survey respondents identified the presence of 57 different substances in SIC 286. Particulates not otherwise regulated and ethylene glycol were estimated to occur most frequently at a total of 222 and 184 processes, respectively. OSHA has retained the existing limit for particulates not otherwise regulated.

SIC 287 - Agricultural Chemicals

This SIC includes establishments that manufacture agricultural chemicals and pesticides. According to the U.S. Department of Commerce, the 1985 value of shipments of SIC 287 ($14.8 billion) represents 7.5 percent of the value of shipments of SIC 28 [2]. The value of shipments in SIC 287 decreased 9.6 percent from 1981 to 1985. Employment in SIC 287 represented 5 percent of SIC 28, but has declined by 16 percent since 1981. The number of establishments in SIC 287 was approximately 9 percent of all establishments in SIC 28 and production workers account for approximately 62 percent of total employment [Table C-1].

SIC 2873 includes manufacturers of nitrogenous and mixed fertilizers. The value of shipments of nitrogenous fertilizers in 1986 was $2.82 billion, a decrease over 1985 shipments [3, p. 13-1].

SIC 2874 includes manufacturers of phosphatic fertilizers, such as phosphoric acid, made from phosphate rock. The value of shipments of phosphatic fertilizers in 1986 was $3.71 billion, which represents a decrease over 1985 shipments [3, p. 13-3]. Ammonia and phosphoric acid are two substances with potential exposure problems that are produced and/or used in SIC 2874.

SIC 2875 includes establishments that mix fertilizers from purchased fertilizer materials. SIC 2879 includes formulators and preparers of ready-to-use agricultural and household pest control chemicals, such as fungicides, insecticides, and herbicides.

In SIC 287, the survey detected more than twice as many small firms (fewer than 20 production workers) as large firms. In small firms, maintenance work is mostly performed by production workers or dedicated maintenance workers. Large firms primarily employ workers specifically for maintenance duties, although some large firms use outside contractors.

The manufacturers classified in SIC 287 usually have two to four basic processes, with numerous firms having potential exposures to as many as five different substances. Thirty-three percent of the processes involve exposure to these chemicals or substances on an intermittent short-term basis (up to 30 minutes), with large firms tending to have longer-term exposures. Thirty-nine percent of the firms in this SIC reported the adoption of an internal exposure standard. Twenty-three percent of the small firms reported the adoption of an internal exposure standard. Nearly 45 percent of the large firms with standards reported using the OSHA PELs; the remainder indicated that they use ACGIH TLVs. The survey found that about 43 percent of the processes are totally enclosed and about 37 percent are located outdoors. Local exhaust ventilation is used most frequently to control exposures at processes not enclosed. About 42 percent of the firms with chemical exposures have respirators for employee use, with a higher percentage of small firms than large firms reporting the presence of respirators. The combined data on exposure levels and methods of exposure control indicate that most plants which are estimated to incur some cost of compliance do not have overexposures at all processes in that plant.

Survey respondents identified the presence of 32 different substances. Ammonia and particulates not otherwise regulated were estimated to occur the most frequently in a total of 344 and 334 processes, respectively. Ammonia was identified in drying/baking, blending/mixing/formulating, packaging/bagging, crushing/grinding/calcining, loading/offloading/receiving/handling, reaction/fermentation, and separation.

SIC 289 - Miscellaneous Chemical Products

This group includes manufacturers of miscellaneous chemical products. For 1985, SIC 289 represented 7 percent ($14.6 billion) of the value of shipments of SIC 28 [2]. From 1981 to 1985, the value of shipments in SIC 289 increased 18.3 percent. The number of employees in SIC 289 was almost 10 percent of SIC 28 and has remained unchanged since 1981. The number of establishments in SIC 289 was approximately 19 percent of all establishments in SIC 28. Production workers equalled approximately 62 percent of all workers [Table C-1].

SIC 2891 includes manufacturers of industrial and household adhesives and sealants. Industry shipments for adhesives and sealants in 1986 amounted to $4.2 billion, of which about 60 percent were by synthetic resins and rubber-based adhesives; 20 percent by sealant and caulking compounds; and the remaining 20 percent by natural-based adhesives and miscellaneous compounds [3, p. 15-3].

SIC 2892 includes manufacturers of explosives, such as TNT (trinitrotoluene). Ethylene glycol dinitrate is one of the products of this SIC which may have potential exposure problems.

SIC 2893 includes manufacturers of printing ink and SIC 2895 includes manufacturers of carbon black. SIC 2899 includes manufacturers of miscellaneous chemical products, not elsewhere classified. Among these three SICs, ethylene glycol, nitrotoluene, hexylene glycol, trimellitic anhydride, and coal dust are all substances with suspected exposure problems that are either produced or used in these sectors.

In SIC 289, the survey identified less than half as many small firms (fewer than 20 production workers) as large firms. In the small firms, maintenance work is performed largely by production workers, whereas large firms primarily rely on a separate maintenance staff.

The manufacturers classified in SIC 289 usually have one to four basic processes, with potential exposure to one to five substances. Forty-seven percent of all employees are exposed to these chemicals or substances on an intermittent basis (up to 30 minutes), with large firms tending to have more long-term exposures. Sixty-five percent of the firms in this SIC reported the adoption of internal exposure standards. Roughly 48 percent of the small firms and 36 percent of the large firms with standards have adopted OSHA PELs. Employee monitoring had been performed at 67 percent of the processes.

The survey found that nearly 29 percent of the processes are totally enclosed and 12 percent are located outdoors. Local exhaust ventilation is used most frequently to control exposures at processes not enclosed. Almost 28 percent of the firms with chemical exposures have respirators for employee use, with a higher percentage of large firms reporting the presence of respirators than small firms. The combined data on exposure levels and methods of exposure control indicate that very few plants which are estimated to incur some cost of compliance have overexposures at all processes in that plant.

Survey respondents identified the presence of 50 different substances in SIC 289. Toluene was estimated to occur the most frequently at a total of 661 processes. Toluene was identified in packaging/bagging blending/mixing/formulating, crushing/grinding/calcining, loading/offloading/receiving/handling, and reaction/fermentation.

SIC 29 - Petroleum and Related Industries

This industry is divided into petroleum refiners and producers of other related products. Petroleum refineries (SIC 2911) produce fuels (such as gasoline, kerosene, and distillate and residual fuel oils) as well as lubricants and chemical feedstocks. These products are produced through straight distillation of crude oil, redistillation of unfinished petroleum derivatives, cracking, or other processes. Other producers in this sector manufacture asphalt and tar products for paving and roofing (SIC 295) and other lubricating oils, greases, and petroleum and coal products (SIC 299) [1, pp. 127-128].

The 1985 value of shipments for SIC 29 ($179.1 billion) was 7.9 percent of the value of shipments for all manufacturing industries. Petroleum refining dominates SIC 29, accounting for 94 percent of this sector's value of shipments [3, pp. 10-8 to 10-14].

The number and size distribution of establishments in SIC 29 are shown in Table C-1, as is total employment. Relative to value of output, SIC 29 has few establishments and low employment, accounting for less than 1 percent of all manufacturing establishments and employment [7, pp. 10, 30].

About 40 percent of the establishments in SIC 29 are petroleum refineries [10], which are large and extremely capital intensive. Production is highly automated; enclosed processes are used throughout. Mean employment size is 105 employees. By contrast, plants in the other industries within SIC 29 are relatively small and less capital intensive, and processes are generally not automated. Mean establishment size in the rest of SIC 29 is 19 employees.

The real value of petroleum product shipments, consumption of petroleum products, petroleum refining capacity, and employment in SIC 29 all peaked between 1977 and 1981. There has been an upturn since 1985, resulting principally from a sharp decline in crude oil prices in the first half of 1986, which stimulated demand for refinery products [3, pp. 10-1 and 10-2]. Demand for petroleum products is expected to grow only slightly in the short run. In the past, trends have been strongly influenced by sharp fluctuations in the price of crude oil [3, pp. 10-3 and 10-4]. In general, low prices for crude oil translate into increased activity for domestic refineries.

The profitability of firms in SIC 29 is low. The median 1985 rate of return on assets (4.4 percent) is the second lowest median return on assets among all 20 two-digit manufacturing industries [6].

Docket comments pertaining to this industry were concerned exclusively with one regulated substance, asphalt. Asphalt is manufactured in petroleum refineries (SIC 2911) and is used to make paving materials (SIC 2951) and roofing materials (SIC 2952). Many commenters [see, for example, Exs. 3-162; 3-420B; 3-895; 3-240; 3-658; 8-5, 581, 3-493B; 3-294; 3-64; 3-22; 3-74; 3-354; 3-966; Tr. 8/9/88, pp. 9-63, 9-65, 9-66, 9-79] provided information on asphalt paving manufacturing, employee exposures, potential costs, and possible impacts; other asphalt applications were not commented on in docket submissions.

Information submitted by firms and trade groups concerned with the manufacture and application of hot-mix asphalt indicated that the manufacture of asphalt paving material falls within SIC 2951, while the activity of paving falls within SIC 1611, Street and Highway Construction. Because the scope of this rulemaking is restricted exclusively to general industry, OSHA has determined that it is most appropriate at this time to defer regulation of asphalt fumes until the Agency has had sufficient time to address the complex health issues associated with this substance and to analyze the impact on the construction industry of establishing a PEL for this substance.

In SIC 29, three out of four firms identified by the survey were small firms (firms with fewer than 20 production workers). In about half of the small firms, maintenance work is performed by production workers; the remainder of small firms employ maintenance workers more often than they use outside contractors for maintenance. Large firms most commonly have a dedicated maintenance staff.

Most employee exposures are intermittent and short-term (up to 30 minutes); of the remaining employee exposures, most are for durations of from 4 to 8 hours (for large firms), or of 1 to 8 hours (for small firms). A slight majority of small firms use some internal exposure standards; most of those that do use internal exposure standards report using OSHA PELs or ACGIH TLVs. Almost 95 percent of larger firms report using internal exposure standards; of these, most report using OSHA PELs, and about one-quarter reported using ACGIH TLVs. Air monitoring data were collected for over half of the processes in large plants, but for less than one-fourth of the processes in small plants.

The survey found that about 30 percent of the processes are totally enclosed and almost two-thirds of plant processes are located outdoors. Production workers use respirators in over 25 percent of processes for firms reporting chemical exposures; however, small firms report a lower percentage of respirator use than do large firms.

Survey respondents identified the presence of 68 different substances in SIC 29. Toluene was estimated to occur the most frequently at a total of 175 processes; trichloroethylene was estimated to occur at a total of 162 processes. Toluene was identified in batch process/coke production and removal, blending/mixing/formulating, and process inspection. Trichloroethylene was identified in blending/mixing/formulating, drying/baking, loading/offloading and measurement.

SIC 30 - Rubber and Miscellaneous Plastics Products Industry

This industry sector consists of establishments that manufacture a variety of products from plastic resins and from natural, synthetic, and reclaimed rubber. Although plastic products account for the largest share of the value of shipments of this industry group, the industry also manufactures a variety of rubber products, including tires, inner tubes, footwear, and belting [1, pp. 129-132]. The value of shipments for 1985 was $71.3 billion. This industry is dominated by the miscellaneous plastic products sector (SIC 307 until 1987 and now SIC 308), which accounts for 81 percent of the establishments, 66 percent of the value of shipments, and 70 percent of the employment for the entire industry group [10]. The tire and inner tube (SIC 301) sector and the miscellaneous rubber products (SIC 306) sector are the other major components of this industry.

Similar processes are used in manufacturing plastic and rubber products, with the nature and form of the final product determining the process more than the product's components. A product's components, however, determine the types of chemical exposures employees experience. Examples of particularly serious types of exposures are those to the foaming agents that are used in the production of foam rubber or plastic foams and to the styrene used to produce polystyrene or in lamination processes.

As shown in Table C-1, this industry sector is characterized by relatively small establishments; 61 percent of establishments have fewer than 20 employees, with an average of 43 employees per establishment. Employment in this industry grew by 7 percent between 1981 and 1985, with growth in the tire and inner tube and miscellaneous plastics product sectors outpacing declines in other sectors [4]. Firms in this industry have above-average profits for manufacturing industries, with a 7.7 percent median rate of return on assets compared with a 7.0 percent median for all manufacturing firms [6].

The only comments received by OSHA that were related to SIC 30, Rubber and Miscellaneous Plastics, concerned the Agency's proposed 50-ppm TWA and 100-ppm STEL limits for styrene [See, for example, Ex. 3-742; Tr. 8/8/88, pp. 95, 177, 178, 180]. Styrene is used in this sector to make a variety of rubber and plastic products, including polyester resins, polystyrene, and a widely used form of artificial rubber. Commenters stated that a small number of the facilities in this sector, i.e., those using styrene resins in open-mold processes, would encounter technological problems in attempting to comply with the proposed styrene limits [Ex. 3-742, pp. 34-36; Tr. 8/3/88, p. 5-95]. This issue is addressed in Chapter F - Technological Feasibility.

Open-mold processes were described by these commenters as operations in which the styrene resin is applied directly to the surface of a mold (generally by means of a spray gun) and is then rolled by hand to build up successive layers of reinforced plastic. When the objects being molded are large, as is the case with boats or underground storage tanks, commenters explained that it is more difficult to position and use local exhaust ventilation effectively [Ex. 3-742, p. 48]. Although most open-mold processes in this sector are involved in the manufacture of plastic bathroom fixtures (showers, tubs, hot tubs, and spas), makers of underground storage tanks and cultured marble products also rely on the open-molding process.

The Styrene Information and Research Council estimates that 265 facilities in this sector use this process to produce bathroom fixtures [Ex. 3-742, p. 105], and the Cultured Marble Institute estimates that a total of 1062 facilities, employing 17,000 workers, manufacture cultured marble products [Tr. 8/3/88, pp. 5-77, 5-177, 5-180]. These firms, like other styrene-using firms in this SIC code, are generally small, privately held firms. The Cultured Marble Institute characterized the typical open-mold-process firm in this sector as a company that employs 17 persons and has annual sales of less than $1 million. The issues of technological feasibility that pertain to users of this process in SIC 30 are discussed in detail in the Technological Feasibility chapter, below.

In this SIC, over 60 percent of the firms identified by the survey were small firms (firms having fewer than 20 production workers). In the small firms, maintenance work is most commonly performed by production workers, although about one-quarter of small firms use outside contractors for maintenance work, and one in seven has a dedicated maintenance staff. Over two-thirds of large firms have dedicated maintenance staff; the remaining large firms use production workers for maintenance more often than they use outside contractors.

Most firms reported using from one to four processes. In SIC 307 (miscellaneous plastics manufacturing), most firms reported using from one to three chemicals, with styrene the most prevalent; however, in rubber manufacturing (SICs 301 to 306), almost half of the firms reported using 6 to 10 chemicals. Most employee exposures in small firms are intermittent and short-term (up to 30 minutes), and there are very few exposures for 4 hours or more. In large firms, by contrast, the majority of chemical exposures are for 4 to 8 hours a day. In this SIC, most small firms have internal exposure standards; the majority of these reported using ACGIH TLVs. Large firms most commonly use OSHA PELs, but many use ACGIH TLVs or have no internal exposure standards. Air monitoring data were provided for about 40 percent of large firms and for approximately 13 percent of small firms.

The survey found that about 37 percent of the processes are totally enclosed, and that very few processes are located outdoors. In one-third of the firms with chemical exposures, production workers use respirators.

Survey respondents identified the presence of 75 different substances in SIC 30. Ethylene glycol was estimated to occur the most frequently at a total of 1,889 processes, including assembly, blending/mixing/formulating, calendaring/winding and coating/spraying. Methyl chloroform was estimated to occur in 1,852 processes including blending/mixing/formulating, coating/spraying, and cutting/sawing/planing.

SIC - Leather and Leather Products

The leather and leather products industry (SIC 31) consists of several sectors such as leather tanning (SIC 311), boot and shoe cut stock (SIC 313), non-rubber footwear (SIC 314), and luggage and leather goods (SICs 315-319) [1, pp. 133-135]. Shipments of leather products increased in 1987, while employment in the leather industry has been declining steadily over the past several years [3, p. 46-1].

Data on the number of establishments and employment for 1985 are shown in Table C-1. In 1985, there were approximately 3,940 establishments engaged in the production of leather and leather products. Over 64 percent of these establishments employed fewer than 20 workers. The largest employer is the non-rubber footwear industry, with 58 percent of the workforce in 1986. Production workers make up 84 percent of the total workforce in SIC 31.

According to the U.S. Department of Commerce, the 1986 value of shipments for leather and leather products ($7.8 billion) was down 8.8 percent from 1985. The total represents 0.4 percent of the value of shipments for all manufacturing industries. Non-rubber footwear (SIC 314) makes up most of the value of shipments in this industry, with 51 percent of the total value [2]. The median return on assets in 1985 for the leather and leather product industry was 6.3 percent [6].

The number of establishments in the leather tanning and finishing industry (SIC 311) has decreased by over 248 establishments, from 384 establishments in 1982 to 136 establishments in 1987. Employment has also decreased significantly while shipments increased to $2.0 billion in 1987 from $1.7 billion in 1986. Since the leather tanning industry is highly dependent on the demand from the non-rubber footwear industry, it is not likely that the situation will improve in the near future [3, pp. 46-1 and 46-2].

The non-rubber footwear industry (SIC 314) had a small increase in the value of shipments in 1987 ($4.1 billion), while total employment and the number of production workers declined 3.0 percent and 2.9 percent, respectively. This industry has suffered substantially since 1981 when an import restraint agreement with South Korea and Taiwan expired. Since then, import's share of the domestic market has increased to over 81 percent in 1987, to an estimated 226 million pairs [3, pp. 46-5 to 46-10]. The miscellaneous luggage and leather goods industry (SICs 315-319) saw improvements in production, employment, and shipments in 1987, reversing a past trend. Shipments were expected to increase 3.9 percent in 1987 to $1.9 billion. The estimated number of production workers also increased in 1987, to 27,200 employees from 27,000 employees in 1986. Imports reached over 52 percent of the domestic market in 1986 [3, pp. 46-10 to 46-14].

In this SIC, the survey identified almost twice as many small firms (fewer than 20 production workers) as large firms. In the small firms, a large share of maintenance work is performed by production workers, although one-fifth of the firms use outside contractors and one-fifth of the firms employ maintenance staffs. Large firms have a dedicated maintenance staff that performs most of the maintenance work, while production workers and outside contractors do the rest of the maintenance work.

The manufacturers classified in this SIC usually have one to three basic processes, with potential exposure to seven to eight chemicals. Employees are exposed to these chemicals on an intermittent short term basis (up to 30 minutes) or continuously (up to 8 hours per day) with large firms tending to have more long-term exposures. Small firms generally have no internal exposure standards. Around one-half of large firms reported using the OSHA PELs; the balance indicated that they have no standards. Air monitoring data was being done at about one-half of the processes found in large plants.

The survey found that over forty percent of the processes are totally enclosed and less than one percent are located outdoors. Local exhaust ventilation is used most frequently to control exposures at processes not enclosed or outdoors. In one-tenth of the firms with chemical exposures, production workers use respirators, with a higher percentage of large firms using respirators than small firms. The combined data on exposure levels and methods of exposure control indicate that overexposure is not occurring at many processes in this SIC.

Survey respondents identified the presence of 42 different substances in SIC 31. N-hexane was estimated to occur the most frequently at a total of 426 processes, primarily gluing/hot pressing. Toluene was estimated to occur at a total of 319 processes including cleaning, coating/spraying, gluing/hot pressing, and stamping/shaping.

SIC 32 - Stone, Clay, and Glass Products

This industry is made up of products such as cement (SIC 324), concrete (SIC 327), pottery (SIC 326), stone (SIC 328), glass (SICs 321-323), and structural clay products (SIC 325). Since these products are primarily used as construction materials, the industry is heavily dependent on the amount of new construction activity in a given year.

There were 21,054 establishments in the stone, clay and glass industry (SIC 32) in 1985. Most of these firms (73.7 percent) employed fewer than 20 workers in 1985. The mean establishment size was 28 employees. Total employment was 514,000 in 1986, a decrease of 1.1 percent over the 1985 total employment figure of 520,000 [7]. Production employment also declined from 1985 to 1986.

In 1986, the value of shipments in SIC 32 ($57.3 billion) increased 3.9 percent over the 1985 figure. The total value was 2.5 percent of the value of shipments for all manufacturing industries. The value of shipments is evenly distributed over the entire industry, except for the concrete sector (SIC 327) with 36.4 percent of shipments [2]. The median rate of return on assets for SIC 32 was 6.5 percent in 1985 [6].

The concrete industry (SIC 327) experienced a small decline in shipments in 1987 after considerable improvement in production, employment, and demand over the past years. The demand for concrete has increased substantially since 1982, when shipments were 23 percent below their current figure. Future demand for concrete depends mainly on non-residential building construction activity [3, pp. 2-7 to 2-8].

The cement industry (SIC 324) experienced a decline in the value of shipments, from $4.1 billion in 1986 to $3.9 billion in 1987, a decrease of 3.3 percent. Consumption of cement also declined in 1987 by 1 percent, the first annual decline since 1982. However, industry shipments were more than 26 percent higher than the 1982 low point of 65 million tons. Total employment was 19,500 in 1987. Production employment (14,500 employees in 1987) represented approximately 74 percent of the workforce [3, pp.2-4 to 2-6].

The glass industry (SICs 321-323) has experienced steady growth over the past two years, mainly in production and shipments. The value of shipments for the glass industry increased from $13.9 billion in 1985 to $14.6 billion in 1986. New product introductions have allowed the glass industry to make substantial gains in winning market share. Total employment and production employment declined for the glass industry (SICs 321-323) in 1986, but SIC 323 (products of purchased glass) did have increases in both total employment and production employment. The outlook for continued growth for the glass industry is good [3, pp. 2-9 to 2-12].

Shipments of structural clay products and pottery (SICs 325-326) have increased substantially over the past few years, from 5.1 billion bricks in 1982 to 7.4 billion bricks in 1986. The 1986 value of shipments for SICs 325-326 was $4.9 billion, an increase of 4.3 percent from 1985. The outlook for the industry is for slow growth in the near future [3, pp. 2-12 to 2-13].

The stone industry (SIC 328) had an increase of 1.3 percent in the value of shipments in 1986. Total employment and production employment stayed virtually the same [2].

In SIC 32, the survey identified over three times as many small firms (fewer than 20 production workers) as large firms. In the small firms, over three-fourths of maintenance work is performed by production workers, although some firms do employ a maintenance staff. Large firms use a dedicated maintenance staff for approximately two-thirds of the maintenance work, while one-fourth use production workers. The remainder of firms use outside contractors.

The manufacturers classified in this SIC usually have one to three basic processes, with potential exposures to as many as eight chemicals. Employees are exposed to these chemicals on an intermittent short term basis (up to 30 minutes) or continuously (up to 8 hours per day), with large firms tending to have more long-term exposures. Small firms generally have no internal exposure standards; when they do, the OSHA PELs are followed about seventy percent of the time. Approximately one-half of large firms reported using the OSHA PELs; the balance indicated that they have no standards or they rely on ACGIH TLVs. Air monitoring data were provided for over one-half of the processes found in large plants.

The survey found that about one-third of the processes are totally enclosed and around one-fifth are located outdoors. Local exhaust ventilation and respirators are used most frequently to control exposures at processes not enclosed or outdoors. In almost one-half of the firms with chemical exposures, production workers use respirators, with a higher percentage of large firms using respirators than small firms. The combined data on exposure level and methods of exposure control indicate that overexposure is not occurring at many processes in this industry.

Survey respondents in this SIC identified blending/mixing/formulating, chipping/grinding, drilling/cutting/flame-jet lancing, polishing (surface) grinding, cutting/sawing/planing, casting, batch making, and bonding as the processes used most often. Chemicals that were present in these processes included: acetone, ammonia, calcium oxide, furfuryl alcohol, graphite, magnesium oxide fume, and silica. The National Lime Association commented on the presence of calcium hydroxide and calcium oxide in this industry [Ex. 3-890].

SIC 33 - Primary Metal Industries

The primary metal industry (SIC 33) is divided into two different sectors: nonferrous metals and foundries (SICs 333-336) and ferrous metals and foundries (SICs 331-332) [1, pp. 145-152]. This includes the basic iron and steel industry, and the metals industry. Both sectors have been hurt in the recent past by a decline in domestic consumption and the growing number of imports into the United States. The future for these industries, however, looks brighter due to an increase in orders, slowing imports, and a decrease in capacity [10]. These industries have had increases in prices, shipments, and profits in 1987 and 1988, helped by the fall in the value of the dollar.

As seen in Table C-1, the number of establishments in SIC 33 in 1985 totaled 10,101. The majority of these firms (55.3 percent) had fewer than 20 employees in 1985. Total employment (808,00 employees in 1985) and production employment (612,000 in 1985) have declined over the last several years, while the average hourly wage of production workers ($12.76 in 1986) has increased by 1.5 percent from 1985 to 1986 [7]. The mean establishment size was 80 employees in 1985. However, according to the American Iron and Steel Institute (AISI), integrated steel mills are typically much larger, averaging 825 workers [Ex. 3-1123, p. 14].

Production in the steel mill products industry has declined over the past few years, from 92.5 million tons in 1984 to 83.0 million tons in 1987, a decline of 10.3 percent [3, p. 20-1]. The 1986 value of shipments ($105.6 billion) in SIC 33 was 4.7 percent of the value of shipments for all manufacturing industries [2]. The median rate of return on assets in 1985 was 5.5 percent for the primary metal industry [6].

In 1987, the industry had its first profitable year since 1982. The industry has cut costs of production while prices have remained steady. In 1988, the industry experienced additional improvement; production was up 15 percent and shipments up 12 percent. Prices and profits rose considerably during the year, and the outlook for 1989 is good [11].

The import situation has also improved for the domestic steel industry, due in part to the falling value of the dollar against major competitors such as Japan and Europe. Imports have been declining since their peak of 26.2 million tons in 1984. Imports as a percent of domestic consumption fell to 22 percent in 1987, down from a peak of 26.4 percent in 1984. Exports reached 1.1 million tons in 1987 [3, pp. 20-1 to 20-9]. Exports during 1988 rose about 50 percent over the previous year.

The ferrous castings industry (SIC 332) has shown a poor performance over the past few years, but is starting to improve. The value of shipments for SIC 332 has increased, from $10.3 billion in 1986 to $10.8 billion in 1987. The value of shipments for SIC 332 is forecast to increase 5.2 percent in 1988, although this trend may not likely to continue in the future. Total employment and the number of production workers have also begun to increase, by 2.3 percent and 1.9 percent, respectively, from 1986 to 1987 [3, pp. 20-6 to 20-7].

Primary nonferrous metals can be classified in four categories:

aluminum, zinc, lead, and copper. Aluminum industry shipments have increased steadily in the past few years, with an 11.3 percent increase in 1987. Shipments are projected to continue rising through 1992 [3, pp.21-8 to 21-11]. Prices during the last several years have continued to increase, from $.53/pound in the last quarter of 1986 to $.83/pound at the end of 1987 and $1.12/pound by the end of 1988 [12,13].

The zinc industry should have steady growth over the next few years, due mainly to an increase in consumption. The price of zinc has risen from $.38/pound in 1986 to $.425/pound in 1987. Domestic consumption increased to 1.014 million metric tons in 1987. The value of shipments increased by 1.4 percent in 1987, and is expected to increase by another 1.6 percent in 1988. Total employment and the number of production workers has remained steady for the past several years [3, pp. 21-14 to 21-16].

Consumption of primary lead products increased slightly over 3 percent in 1987-88 owing to increases in the replacement battery market. Automotive products account for about 70 percent of all demand for lead. Changes in recycling patterns due to EPA OKRA regulations may increase demand for primary lead in the near future. The market in general has been growing at a steady 1 percent per year [3, pp. 21-6 to 21-8 and 14]. Prices have risen in recent months to $.42/pound [12] from $.369 in 1987. ASARCO, one of two primary lead producers in the U.S., is considering adopting London Metal Exchange prices in lieu of its own pricing [4].

The copper industry has been undergoing restructuring to remain competitive in the world market. Currently, there are seven operating copper smelters, compared to fourteen in 1970. This restructuring has forced the industry to decrease capacity and reduce employment [3, pp. 21-11 to 21-14]. The price of copper has increased from $.661/pound in 1986 to $.750/pound in 1987 due to a decline in inventories [3, pp. 21-11 to 21-14]. Current 1988 cash prices for copper have risen to $1.64/pound [12]. The Peruvian copper fields are estimated to need an additional 30 days to return to full production following the recent 54 day strike by miners [15] This should allow the industry to turn a profit for the first time in several years. The copper smelting industry is likely to be impacted by the proposed revision to the PEL for sulfur dioxide.

In SIC 33, the survey identified slightly more small firms (fewer than 20 production workers) than large firms. Maintenance work in the small firm is done primarily by production workers although some firms use a dedicated maintenance staff. Large firms generally have maintenance work performed by the maintenance staff, with the remainder of firms using production workers and outside contractors.

The manufacturers classified in this SIC who reported chemical or process use usually have one to four basic processes, with potential exposure to one to four chemicals. Employees are exposed to these chemicals on an intermittent short term basis (up to 30 minutes) or continuously (up to 8 hours per day), with large firms tending to have more long-term exposures. Small firms generally have some internal exposure standards; when they do, the OSHA PELs are followed about three-fourths of the time. Over one-half of large firms reported using the OSHA PELs; the balance indicated that they have no standards or they rely on ACGIH TLVs. Air monitoring data were provided for approximately two-thirds of the processes found in large plants.

The survey found that about one-fourth of the processes are totally enclosed and less than 3 percent are located outdoors. Local exhaust ventilation and respirators are used most frequently to control exposures at processes not enclosed or outdoors. In almost one-half of the firms with chemical exposures, production workers use respirators, with a higher percentage of large firms using respirators than small firms. The combined data on exposure levels and methods of exposure control indicate that overexposure may occur at less than one-tenth of the processes in small firms and at about one-fifth of the processes in large firms.

Survey respondents in SIC 33 identified metal melting/pouring/casting as the process most frequently used, with exposure to aluminum metals, carbon monoxide, and copper fume reported most frequently. The American Cast Metals Association confirmed the presence of most of the chemicals surveyed [Exs. 3-673 and 3-675]. The American Iron and Steel Institute also commented on several of the chemicals identified in the survey [See, for example, Ex. 3-1123].

SIC 34 - Fabricated Metal Products

The fabricated metal products industry (SIC 34) can be broken down into nine categories: metal cans and shipping containers (SIC 341); cutlery and hand tools (SIC 342); heating equipment (SIC 343); fabricated structural metal products (SIC 344); screw machine products, bolts, and washers (SIC 345); forgings and stampings (SIC 346); plating and coating (SIC 347); small arms and ordnance (SIC 348); and miscellaneous wire and fabricated products (SIC 349). SIC 34 excludes machinery and transportation equipment [1, pp. 153-166].

The total number of establishments in the fabricated metal products industry in 1985 was 46,322. The majority of these firms (67.0 percent) have fewer than 20 employees, a change of 0.2 percent since 1984. Total employment in this industry has reached 1.5 million employees, an increase of 0.1 percent since 1984 [7].

The 1986 value of shipments for SIC 34 ($138.0 billion) represents a 1.1 percent decrease over 1985. This was 6.1 percent of the value of shipments for all manufacturing industries [2]. The median return on assets for the fabricated metal products industry in 1985 was 7.1 percent [2].

Metal can (SIC 3411) shipments have been increasing steadily in the past few years, from 104.7 billion units in 1986 to 109.3 billion units in 1987, an increase of over 4.4 percent. This was due mainly to the increase in soft drink and beer cans being shipped. The value of shipments has also increased, with a compound annual increase of 2.9 percent from 1980 to 1985. Total employment in the metal cans industry has remained steady, with a slight increase expected in 1987. The number of production workers has increased slightly, with an increase of 0.3 percent from 1986 to 1987. Exports of metal cans have decreased substantially since 1984 when they reached an all-time high of $56.5 million. Since that time they have decreased to $36.2 million in 1987 [3, pp. 7-1 to 7.4].

The fabricated structural metal industry (SIC 3441) produces structural metal components used primarily in the construction industry. Shipments of fabricated structural metal decreased slightly, from $9.0 billion in 1986 to $8.9 billion in 1987. Total employment decreased slightly in 1987 [3, pp. 2-3 to 2-5].

The value of shipments in the screw machine products, bolts, and washers industry (SIC 345) decreased slightly from 1986 to 1987, from $7.8 billion to $7.9 billion. Total employment increased from 94,000 in 1986 to 94,400 in 1987. Since the automotive industry is the major customer for this industry, stable automotive sales are the key to economic health for this industry sector [3, pp. 26-1 to 26-6].

In SIC 34, the survey identified over twice as many small firms (fewer than 20 production workers) as large firms. In the small firms, about one-half of the firms have maintenance work performed by production workers, the remaining firms using maintenance workers or outside contractors. Large firms generally employ a maintenance staff to do the majority of maintenance work, although some firms use production workers and outside contractors.

The manufacturers classified in this SIC usually have one to three basic processes, with potential exposure to one to four chemicals. Employees are exposed to these chemicals on an intermittent short term basis (up to 30 minutes) or continuously (up to 8 hours per day) with large firms tending to have more long-term exposures. Small firms generally have no internal exposure standards; when they do, the OSHA PELs are followed about one-half of the time. Over one-half of large firms reported using the OSHA PELs; the balance indicated that they have no standards or they rely on ACGIH TLVs. Air monitoring data were provided for about one-half of the processes found in large plants.

The survey found that about one-fourth of the processes are totally enclosed and around one-fifth are located outdoors. Local exhaust ventilation and respirators are used most frequently to control exposures at processes not enclosed or outdoors. In over one-half of the firms with chemical exposures, production workers use respirators, with a higher percentage of large firms using respirators than small firms. The combined data on exposure levels and methods of exposure control indicate that overexposure is not occurring at any processes in small firms and at less than one-tenth of the processes in large firms.

Survey respondents in this SIC identified casting/painting, welding/soldering, polishing (surface)/grinding, and degreasing, as the processes most frequently used. Welding fumes, iron oxide, and isopropyl alcohol were the chemicals identified most often in the survey. OSHA has retained the existing limit for iron oxide. No comments were received relative to processes and chemicals in this SIC.

SIC 35 - Non -Electrical Machinery

The non-electrical machinery industry (SIC 35) is made up of several different sectors: engines and turbines (SIC 351); farm and garden machinery (SIC 352); construction and related machinery (SIC 353); metalworking machinery (SIC 354); special industry machinery (SIC 355); general industrial machinery (SIC 356); computer and office equipment (SIC 357); refrigeration and service industry machinery (SIC 358); and miscellaneous machinery and equipment (SIC 359) [1, pp. 167-183].

As seen in Table C-1, the number of establishments in 1985 totaled 77,748. The majority of these (77.1 percent) had fewer than 20 employees in 1985. Total employment and production employment have decreased over the last several years. The 1986 value of shipment ($208.5 billion) in SIC 35 was 9.2 percent of the value of shipments for all manufacturing industries [2]. In 1985, the median rate of return on assets for SIC 35 was 7.5 percent [6].

The 1986 value of shipments for engines and turbines (SIC 351) was $14.1 billion, a decrease of 5.5 percent of the 1985 value of shipments ($14.9 billion). Both total employment and production employment decreased from 1985 to 1986, by 8.8 percent and 9.3 percent, respectively. Major expansions of electrical power generation capacity and hence, turbine manufacture have been curtailed in recent years as cogeneration facilities are now providing additional power . Smaller units for these same cogeneration facilities have provided some additional sales [16]. The largest sector of SIC 351 is internal combustion engines, n.e.c., with 77 percent of the value of shipments in 1986.

The farm and garden machinery industry has experienced some improvement in 1987. While the value of shipments for lawn and garden equipment increased in 1987, the value of shipments for farm machinery and equipment ($7.0 billion) declined to their lowest level since 1973. Total employment, which exceeded 125,000 in 1981, dropped to around 67,000 in 1987. Production employment, which makes up approximately two-thirds of the work force, has also been declining since 1979. The prospects for lawn and garden equipment appear much better, with steady increases in the value of shipments since 1981. The 1987 value of shipments for lawn and garden equipment ($3.7 billion) was 4.2 percent greater than the 1986 value of shipments ($3.5 billion) [2]. This industry had a compound annual increase of 8.9 percent from 1980 to 1985 in value of shipments. Total employment and production employment have remained fairly steady, with compound annual increases of 1.9 percent and 3.8 percent, respectively [3, pp. 25-1 to 25-3].

The construction and related machinery industry (SIC 353) has experienced a decline in recent years. The value of shipments for SIC 353 declined by 6.2 percent, from $27.7 billion in 1985 to $25.9 billion in 1986. Both total employment and production employment fell from 1985 to 1986, by 8.7 percent and 11.7 percent, respectively [2]. The decline of the dollar value of shipments must be viewed against a background of reorganization and price cutting by American manufacturers resulting in leaner, more efficient organizations that can make a profit at lower levels of sales. Significant market share has been regained [17]. Construction machinery makes up the largest share of this industry, with approximately half of the total value of shipments.

The machine tool industry has had a major improvement in orders, and profits during 1988. "Orders for all of 1988 climbed to about $3.5 billion, up 66% from the $2.1 billion range for both 1986 and 1987" [18]. The 1986 value of shipments for metal working machinery (SIC 354) was $20.5 billion, an increase of 3.2 percent over 1985. Although shipments increased in this industry in 1986, both total employment and production employment fell during the same time period [2]. This is a reflection of the downsizing and modernizing that has been undertaken in this industry. In the future, moderate sales improvements should have a positive impact on earnings [19]. The largest sector within the metalworking industry is special dies, tools, jigs, and fixtures, with 38 percent of the value of shipments and 43 percent of the total workforce.

Special industry machinery (SIC 355) has experienced stable growth in the past, and this trend is likely to continue into the future. Industry shipments increased approximately one percent, from $14.8 billion in 1985 to $14.9 billion in 1986. Special industry machinery, n.e.c. (SIC 3559) is the largest sector within this industry, with 41.9 percent of the total value of shipments in 1986. Both total employment and production employment have been falling, by 5.0 percent and 7.5 percent, respectively, from 1985 to 1986. Production employees make up approximately 57.0 percent of the total workforce.

The 1986 value of shipments for general industrial machinery, SIC 356, ($24.8 billion) fell from the 1985 value of shipments ($25.3 billion) by an estimated 2.4 percent. Employment and industry shipments are divided fairly evenly over the entire industry, with pumps and pumping equipment (SIC 3561) and general industrial machinery, n.e.c. (SIC 3569) being the largest sectors. Total employment declined by 4.1 percent, while the number of production workers fell by 5.8 percent [4].

The computer industry (SIC 357) has had stable demand for its products in the U.S. market during 1986 and 1987. The value of shipments of office and computing machines (SIC 357) decreased from $62.2 billion in 1985 to $58.8 billion in 1986, a decline of 5.5 percent reflecting strong price competition. Electronic computing equipment (SIC 3573) is the largest segment, with 89 percent of the value of shipments. Total employment and the number of production workers have declined since 1985 by 10.8 percent and 12.7 percent, respectively. Imported computer equipment has made significant inroads into the domestic market, due mainly to the standardization of products and the fall in the price of computer equipment [3, pp. 30-1 to 30-11]. The dollar value of 1988 shipments is ahead of 1987 shipments [20].

The refrigeration and service machinery industry (SIC 358) had an annual rate of growth of 0.9 percent from 1985 to 1986, attributable to the increase in new residential construction. While total employment and the number of production workers have increased, imports have also been steadily increasing [3, pp. 22-9 to 22-11].

It appears that the general industrial machinery industry (SIC 35) could be affected by several of the proposed revisions. The following substances are used or generated by this industry: carbon dioxide, chlorine, chromium metal, fibrous glass dust, furfuryl alcohol, iron oxide, manganese fumes, nitrogen dioxide, oil mist, sulfur dioxide, 1,1,2-trichloro-1,2,2-trifluoroethane, triethylamine, tungsten, welding fumes, wood dust and asphalt fumes. The majority of comments from the general industrial machinery industry deal with the appropriateness of the PELs rather than technical or economic feasibility.

The Association of Reproduction Materials Manufacturers (ARMM) commented on their opposition to the proposed revision for ammonia based on health effects and the inappropriateness of adopting ACGIH standards. ARMM is a trade group with 47 company members who supply materials and equipment to over 5,000 commercial blueprinters [Ex. 8-29].

The International Institute of Ammonia Refrigeration opposed the proposed standard for ammonia based on health effects and economic feasibility. In the final rule, only a STEL of 35 ppm has been set for this substance.

In this SIC, the survey identified over four times as many small firms (fewer than 20 production workers) as large firms. In the small firms, maintenance work is performed in large part by production workers, although some firms employ dedicated maintenance staffs or use outside contractors. Large firms have the majority of maintenance work performed by a dedicated maintenance staff, with some use of production workers or outside contractors.

The manufacturers classified in this SIC usually have from one to four basic processes, with potential exposure to one to four chemicals. Employees in SIC 35 are exposed to these chemicals for varying amounts of time from intermittent short term periods (up to 30 minutes) to continuously (up to 8 hours per day), with small firms having more intermittent short term exposures and large firms tending to have more long-term exposures. Small firms generally have no internal exposure standards; when they do, the OSHA PELs are followed about one-third of the time. Over one-half of large firms reported using the OSHA PELs; the balance indicated that they have no standards or they rely on ACGIH TLVs. Air monitoring data were provided for about one-half of the processes found in large plants.

The survey found that about one-fifth of the processes are totally enclosed and 5 percent are located outdoors. Local exhaust ventilation, general dilution ventilation and respirators are used most frequently to control exposures at processes not enclosed or outdoors. In over one-half of the firms with chemical exposures, production workers use respirators, with large firms and small firms using respirators at about the same rate.

Survey respondents in this SIC identified polishing (surface)/grinding, coating/painting, and soldering as the processes which occur most frequently. Chemicals that were present most often were welding fumes, oil mist, and stoddard solvent. Comments from Caterpillar Incorporated and John Deere and Co. confirmed the presence of several of the survey chemicals in SIC 35 such as chromium metal, iron oxide, oil mist, welding fumes, and 1,1,2-trichloro-1,2,2-trifluoroethane [Ex. 3-349]. In the final rule, OSHA has not revised the existing limits for chromium metal, iron oxide and oil mist.

SIC 36 - Electric and Electronic Equipment

This industry is made up of several distinct sectors: electric distributing equipment (SIC 361); electrical industrial apparatus (SIC 362); household appliances (SIC 363); electrical lighting and wiring equipment (SIC 364); radio and TV receiving equipment and communication equipment (SIC 365-366); electronic components and accessories (SIC 367); and miscellaneous electronic equipment (SIC 369) [1, pp. 184-195].

In 1985, the electric and electronic equipment industry employed about 2.2 million workers. The majority of the firms had fewer than 20 employees. The value of shipments for all electric and electronic equipment establishments in 1986 was $196.2 billion. This was 8.7 percent of the value of shipments for all manufacturing industries. Most of the value of shipments in SIC 36 is from the communication equipment sector ($67.4 billion or 34.4 percent) [2]. The median return on assets for the electric and electronic equipment industry was 7.9 percent in 1985 [6].

The electric distributing equipment industry (SIC 361) had mixed performance during 1987. While the value of shipments increased for switchgear by 0.5 percent, the value of shipments for transformers decreased by 6.6 percent from 1986 to 1987. Total employment and the number of production workers has remained fairly steady since the early 1980's [3, p. 28-4].

Motors and industrial controls (SIC 362) have had stable sales during the last several years. Future growth is dependent upon the economy in general and construction growth for any sizable increases in sales. Motors have significant import pressure; several domestic manufacturers have manufacturing facilities in Mexico. Industrial controls are expected to grow by 2.5 percent [3, pp. 28-1 to 28-3].

The household appliance industry (SIC 363) has had a steady increase in shipments since the early 1980's, from $16.8 billion in 1986 to $17.7 billion in 1987, an increase of 5.7 percent. The industry is optimistic about its future, due mainly to increased residential construction and an anticipated increase in disposable income. Imports have not been a substantial factor in this industry (exports have not increased either). Total employment and the number of production workers declined from 1980 to 1985 by 4.4 percent and 4.3 percent, respectively. This decline in employment is due to the recent number of acquisitions within the industry and the need to cut costs of production. In 1987, total employment and production employment increased slightly [3, pp. 47-8 to 47-11].

The value of shipments for the electrical lighting and wiring industry (SIC 364) has been increasing steadily over the last decade, from 11,321 in 1980 to 15,806 in 1985, an increase of 39.6 percent. However, total employment and the number of production workers have decreased slightly. Performance in this industry is related, in part, to activity in the construction industry. Since the electrical lighting and wiring industry depends on both residential and non-residential construction, it is able to withstand a slowdown in one sector as long as the other sector is still active [3, pp. 4-1 to 4-4].

The consumer electronics and communication equipment industry (SICs 365-366) has had mixed performance in the past. The communication equipment industry has performed well, while the consumer electronics industry has not performed as well, due to import competition. Overall, the value of industry shipments has remained fairly stable, with shipments increasing in the communication equipment industry and shipments decreasing in the consumer electronic industry. Total employment and the number of production workers also follow this pattern, decreasing for consumer electronics and increasing for communication [3, pp. 31-1 to 31-8 and 32-1 to 32-6 and 47-7].

The electronic components and accessories industry (SIC 367) is expected to show record growth over the next few years. Industry shipments were up 8.3 percent, from $43.9 billion in 1986 to $47.5 billion in 1987. This was due, in part, to the strong performance of the defense electronics industry. The number of production workers and total employment have remained fairly steady in 1986 and 1987. Imports are still increasing, but may be slowed due to the fall in the value of the dollar [2, pp. 32-1 to 32-4].

In SIC 36, the survey estimated that almost 70 percent of the firms are small firms (fewer than 20 production workers). Maintenance work is usually performed by production workers in the small firms and a dedicated maintenance staff for the large firms.

The manufacturers classified in this SIC usually have one to three processes, with potential exposure to one to six chemicals. Employees are exposed to these chemicals on an intermittent short term basis (up to 30 minutes) or continuously (up to 8 hours per day), with large firms tending to have more long-term exposures. Small firms generally have no internal exposure standards; when they do, the OSHA PELs are followed three-fourths of the time. Almost two-thirds of large firms reported using the OSHA PELs; the balance indicated that they use the ACGIH TLVs most frequently. Air monitoring data were provided for almost one-half of the processes found in large plants.

The survey found that less than one-fifth of the processes were totally enclosed and less than one percent located outdoors. General dilution and local exhaust ventilation are used about equally to control exposures at processes not enclosed or outdoors. In about one-quarter of the firms with chemical exposures, production workers use respirators, with a higher percentage of large firms using respirators than small firms. The combined data on exposure levels and methods of exposure indicate that overexposures may occur only in some processes in this industry.

Survey respondents in this SIC identified coating/painting, polishing (surface)/grinding, processing, and degreasing as the processes which occur most frequently, and tin, stoddard solvent, and zinc oxide as the chemicals most frequently used. No commenters addressed the processes or chemicals in this SIC.

SIC 37 - Transportation Equipment

This industry sector includes establishments engaged in manufacturing equipment for land, sea, air, or space transportation and includes manufacturers of parts and accessories as well as complete vehicles.

The major subdivisions within this sector are motor vehicles and motor vehicle equipment (SIC 371), aircraft and parts (SIC 372), ship and boat building and repair (SIC 373), railroad equipment (SIC 374), motorcycles, bicycles and parts (SIC 375), guided missiles, space vehicles and parts (SIC 376), and miscellaneous transportation equipment (SIC 379). Establishments in the miscellaneous subdivision manufacture a broad range of products (e.g., from tanks to wheelbarrows) [1, pp. 196-201]. Because the manufacture of transportation equipment involves a wide range of industrial processes, establishments in this sector often include or involve foundries, electroplating operations, various types of hot metal work, welding, laminating, plastic molding, and painting and coating. Workers may be exposed to many chemicals used in these processes.

Although the transportation equipment industry includes both very small and very large establishments, it has an unusual number of very large establishments employing thousands of employees. These very large establishments are most likely to be found in plants that produce final equipment on a mass-production basis (e.g., automobile plants, aircraft plants, tank assembly lines). However, as shown in Table C-1, 68 percent of all establishments have fewer than 20 employees.

The prosperity of the industry fluctuates with business cycles and with the value of the dollar. Employment in this industry declined between 1981 and 1982 but had recovered to the 1981 level by 1984 and had increased another 4 percent by 1985 [2].

The record contains comments from businesses which use styrene in open-molding processes to produce reinforced plastics products such as fiberglass boats, fiberglass car and truck bodies, and transportation equipment parts [Ex. 3-742, pp. 34-36; Tr. 8/3/88, pp. 5-95, 5-119]. Commenters noted that controlling employee exposures during the open molding of large components (e.g., boat hulls and decks, recreational vehicles) is made costly and difficult by the large sizes and bulky configurations of these products [Ex. 3-742, p. 48].

The open-mold process in this sector is similar to that in other reinforced plastics industries in that it involves the use of a styrene resin to make a mold, followed by the application of a fiberglass-styrene-catalyst mixture with a spray or "chopper" gun, followed by manual rolling of the recently applied surface. Workers bend over the mold to perform the layup operation, which requires rolling with a short or long-handled roller. The roller, spray gun, and other tools used in this process all require repeated cleaning with acetone in order to operate smoothly, and the workers themselves use acetone at frequent intervals to clean the styrene resin from their skin.

The Styrene Information and Research Council (SIRC) estimates that there are 625 reinforced plastic boatmakers in this sector that produce boats under 30 feet in length, and 125 facilities that manufacture larger boats [Ex. 3-742, p. 105]. These boatmakers are estimated to employ about 32,000 production workers. However, SIRC estimates that no more than 20 percent of these employees engage in open molding or work in portions of these facilities where such molding is being done [Tr. 8/3/88, p. 5-100].

Most boat builders are small firms, and many are family-owned enterprises with only one facility. Because the purchase of a recreational boat is a discretionary expense, the industry is relatively price-sensitive. For example, Jeff Napier, president of the National Marine Manufacturers Association, stated that the price elasticity of boat sales was approximately 2, i.e., every 1-percent increase in price results in a 2-percent decline in sales [Tr. 8/3/88, pp. 5-168, 5-169]. The boat building industry is currently undergoing expansion and is enjoying relatively high profits [Tr. 8/10/88, pp. 10-144, 10-145]. Boat building is a labor-intensive industry, and firms in this sector argue that automation is not an option, since many recreational boats are custom designed [Tr. 8/10/88, pp. 10-144, 10-145].

In this SIC, the survey identified twice as many small firms (fewer than 20 production workers) as large firms. In the small firms, maintenance work is performed for the most part by production workers, although some firms use outside contractors or have a dedicated maintenance staff. Large firms divide maintenance work about equally between a dedicated maintenance staff and production workers.

The manufacturers classified in this SIC usually have two to four basic processes, with potential exposure to one to six chemicals being most common, though some firms report using up to ten chemicals. Employees are exposed to these chemicals on an intermittent short-term basis (up to 30 minutes) or continuously (up to 8 hours per day) with large firms tending to have more long-term exposures. The majority of small firms use the OSHA PELs. Over 60 percent of large firms reported using the OSHA PELs; the balance had no standards. Air monitoring data was being done for about one-half of the processes found in large plants.

The survey found that about 40 percent of the processes are totally enclosed and between 5 and 10 percent are located outdoors. Local exhaust ventilation and respirators are used most frequently to control exposures at processes not enclosed or outdoors. In over 70 percent of the firms with chemical exposures, production workers use respirators, with a higher percentage of small firms using respirators than large firms. The combined data on exposure levels and methods of exposure control indicate that overexposures may occur at all processes in small firms and at about one-half of the processes in large firms.

Survey respondents identified the presence of 68 different substances in SIC 37. Welding fumes was estimated to occur the most frequently at a total of 3,508 processes, toluene at 3,191, and styrene at 2,541. Welding fumes were identified in machining/grinding, welding/brazing, coating/spraying, and materials manufacture/fabrication. Toluene was identified in adhesive binding, assembly, coating/spraying and cutting/sawing. Styrene was identified in injection molding, coating/spraying, sanding and assembly.

SIC 38 - Measuring, Analyzint and Controlling Instruments

SIC 38 includes manufacturers of instruments used to measure, test, analyze and control. It also includes optical instruments and lenses; surveying and drafting instruments; hydrological, hydrographic, meteorological, and geophysical equipment; search, detection, navigation, and guidance systems and equipment; surgical, medical, and dental instruments, equipment, and supplies; and watches and clocks [1, p. 243].

The industries in this SIC rely heavily on research and development activities (R&D) of other industries for sales of their products. According to the U.S. Department of Commerce, increases in research and development expenditures by industry and government in 1986 caused increases in sales of scientific and industrial instruments [2]. High tech firms, which represent a large portion of SIC 38's product market, are the largest investors in research and development, where R&D expenditures are measured as a percentage of gross sales. Firms producing semiconductors, computers and related equipment, office equipment, and software, among others, were major sources of R&D funds in 1986. The pharmaceutical and chemical industries also have relied on R&D to a large extent. In addition, the decline in the price of oil, which raises profits by lowering production costs, is expected to further stimulate R&D expenditures by the chemical industry [3, p. 33-1].

Similarly, government outlays for R&D increased in 1986 by more than 9 percent in current dollars. Most of the R&D expenditures, however, were for defense-related research. In addition, the National Aeronautics and Space Administration (NASA) is expected to invest in new instrumentation for the redesign of the space shuttle and other rocket systems [3, p. 33-4].

According to the U.S. Department of Commerce, the value of shipments in 1985 ($61 billion) increased almost 26 percent since 1981 [2]. Between 1981 and 1985, SIC 38 experienced a 1 percent loss in employment [3]. Of all employees working in SIC 38 , 54.4 percent were production workers [5]. In 1985, the median rate of return on assets in this SIC was 7.3 percent [6].

From 1981 to 1985, the value of shipments for SICs 383 and 384 experienced growth, rising 60 and 54.3 percent, respectively. SIC 383 comprises 8 percent of the total value of shipments in SIC 38, while SIC 384 represents 23 percent. In contrast, SIC 387 experienced a drop of 36 percent in the value of shipments, representing only 1.5 percent of the total value of shipments in SIC 38 [6].

In SIC 38, the survey identified nearly twice as many small firms (fewer than 20 production workers) as large firms. In both small and large firms, maintenance work is performed predominantly by workers specifically employed to handle maintenance duties.

The manufacturers classified in this SIC usually have two to six basic processes, with potential exposures to as many as six substances. Fourteen percent of the processes involve exposure to these chemicals or substances on a short-term basis (up to 30 minutes), with small firms tending to have shorter-term exposures. Fifty percent of the firms in SIC 38 have reported the adoption of internal monitoring standards. Of those firms with standards, the most frequently reported were OSHA PELs. Employee monitoring had been performed at 32 percent of the processes. The survey found that about 49 percent of the processes are totally enclosed and 10 percent of the processes are located outdoors. Local exhaust ventilation is used most frequently to control exposures at processes not enclosed. Nearly 45 percent of the firms with chemical exposures have respirators for employee use, with a higher percentage of small firms reporting the presence of respirators than large firms. The combined data on exposure levels and methods of exposure control indicate that many plants which are estimated to incur some cost of compliance have overexposures in all processes at that plant.

Survey respondents identified the presence of 30 different substances in SIC 38. Isopropyl alcohol was estimated to occur the most frequently at a total of 2,294 processes. Isopropyl alcohol was identified in blending/mixing/formulating, adhesive binding, recovery/reprocessing/ reclamation, drying/baking, packaging/bagging, extrusion, loading/ offloading/receiving/handling, reaction/fermentation, boilers, coating/spraying/finishing/layup, separation, and crushing/grinding/ calcining.

SIC 39 - Miscellaneous Manufacturing Industries

Miscellaneous industries included in SIC 39 reflect a diverse group of producers. Most of the industries in SIC 39 produce discretionary durable consumer goods, some of which are luxury goods. Establishments that cannot be grouped together at the three-digit level are included in SIC 399. At the three-digit level, miscellaneous manufacturing industries include producers of jewelry, silverware, and plated ware (SIC 391); musical instruments (SIC 393); toys and sporting goods (SIC 394); pens, pencils, office and art supplies (SIC 394); and costume jewelry and notions (SIC 396). A sixth category, miscellaneous manufactures (SIC 399), includes producers of brooms and brushes, signs and advertising displays, burial caskets, hard surface floor coverings, and manufacturing industries "not elsewhere classified" [1, pp. 211-218].

The number of establishments and employment in SIC 39 are shown in Table C-1. Nearly three-quarters (72 percent) of these employees are production workers.

Establishments in SIC 39 are generally smaller than those in manufacturing as a whole, with higher proportions of employees concentrated in small establishments. The mean size of establishments is 11 employees, with 85 percent of establishments having fewer than 20 employees, compared with less than 65 percent for manufacturing establishments as a whole. Relatively few establishments in SIC 39 have 100 or more employees [7].

Miscellaneous manufactures (SIC 399) has the largest share (more than one-third) of the value of shipments for SIC 39 ($26.5 billion in 1985) [2, Vol. 1: 8, 22, 24]. Substantial import competition, however, poses a threat to various subsectors. Imports account for nearly 60 percent of the new supply of sporting and athletic goods and between one-quarter and three-eighths of new supply in many other industries. The recent decline of the dollar has tended to halt or reverse import penetration [3, pp. 45-2 to 45-11; 46-10 to 46-13]; however, domestic production in SIC 39 will be affected by the trend among doll and toy manufactures to move offshore [3, pp. 45-2 to 45-11; 46-10 to 46-12].

In terms of profitability, the majority of industries in SIC 39 are more profitable than most manufacturing industries. The median 1985 rate of return on assets (8.0 percent) is the second highest median return on assets of all two-digit manufacturing industries. Median rates of return for four-digit industries within this sector range from 3.4 percent to 9.5 percent [6].

In this SIC, the survey found that small firms (fewer than 20 production workers) comprise 85 percent of the total number of firms. In the small firms, maintenance work is performed for the most part by production workers, where large firms have a dedicated maintenance staff.

The manufacturers classified in this SIC usually have two or three basic processes, with potential exposure to two chemicals. Employees in small firms are exposed to these chemicals about equally either on a short-term intermittent basis (up to 30 minutes) or continuously (up to 8 hours per day) with somewhat fewer employees exposed for periods between these two extremes. Large firms have more long-term exposures. Small firms generally either have no internal exposure standards, or use the OSHA PELs, with some using the ACGIH TLVs. About one-half of the large firms use the OSHA PELs; most of the balance indicated that they have no standards. Air monitoring data were provided for one-fifth of the processes found in small firms, and over one-third in large firms.

The survey found that about one-quarter of the processes are totally enclosed and 20 percent are located outdoors. Local exhaust ventilation is used most frequently to control exposures at processes not enclosed or outdoors. In almost one-half of the firms with chemical exposures, production workers use respirators, with a higher percentage of large firms using respirators than small firms. The combined data on exposure levels and methods of exposure control indicate that most firms will have no processes where overexposures may occur.

Survey respondents in this SIC identified gluing/hot pressing, coating/spraying/finishing/layup, and cutting/sawing/planing as the processes most frequently used. Stoddard solvent, toluene, particulates not otherwise regulated and styrene are the substances encountered most often.

The Casket Manufacturers Association of America (CMAA) [Ex. 8-78], representing firms in SIC 3995, Burial Caskets, submitted information to the docket describing the manufacturing processes and material used by facilities in this four-digit sector. According to the CMAA, the primary materials of construction of caskets are hardwood, metal, and cloth-covered board. Firms in the hardwood segment of this industry expressed concern that the proposed limit for hardwood of 1 mg/m(3) would require the installation of controls and the imposition of compliance costs. (OSHA notes that the final rule's PEL for hardwood dust is 5 mg/m 3.) The CMAA reports that there are about 20 companies that assemble hardwood caskets; two of these firms account for more than half of the total unit volume of production [Ex. 8-78]. Most firms in this segment have less than $5 million in annual sales, although larger firms have $15 million in sales annually [Ex. 8-78]. According to Robert Morris Associates' financial data for SIC 3995, after-tax profits in this sector are $7,900,000.

No comments representing firms in SIC codes other than 3995 submitted industry profile information to the docket; however, the sporting goods manufacturers (SIC 3949) submitted cost and feasibility data to OSHA, and these data are discussed in the Technological Feasibility and Costs of Compliance chapters, below.

SIC 40 - Railroad Transporatation

SIC 40 includes establishments that provide line-haul railroad transportation, and switching and terminal establishments. General authority for the working conditions at railroad operations is vested in the Federal Railroad Administration. For the most part, OSHA's standards apply only to off-track operations such as shops and servicing areas. The U.S. Department of Commerce estimates that in 1987, there were 23 individual Class I railroads (those with operating revenues of $88.5 million or more in 1986 dollars), which accounted for over 90 percent of the freight tonnage handled by the railroad industry [4]. The industry also includes about 480 smaller carriers, including shortlines and switching and terminal companies. The 1987 operating revenue for the railroad industry was estimated at $26.5 billion, representing a gain of 1.1 percent over 1986. Revenue ton-miles were estimated as 930 billion, which represents more than a 7 percent rate of growth [3, p. 55-8]. Between 1980 and 1985, the industries in SIC 40 experienced a serious economic decline, as indicated by the fact that SIC 40 was the second slowest growing SIC (behind SIC 10, metal mining), and third highest in terms of employment losses (behind SIC 33, primary metals and SIC 35, heavy machinery). During this period, employment declined by approximately 27 percent [3, pp. 13-14]. The median rate of return on assets in 1985 was 4.4 percent [6].

In SIC 40, establishments generally employ dedicated maintenance workers or hire an outside firm to perform maintenance functions. The establishments in this SIC generally have one or two basic processes with potential exposure to two or three substances. Over 80 percent of the processes involve exposure to these chemicals or substances on an intermittent short-term basis (up to 30 minutes). Fifty percent of the establishments in SIC 40 have reported the adoption of an internal exposure standards. Among those establishments with internal standards, most use the ACGIH TLVs.

The survey found that none of the interviewed establishments had totally enclosed processes, but 85 percent were located outdoors. Of those establishments with ventilation systems, all were locally exhausted. None of the respondents reported having respirators available in processes. The combined data on exposure levels and methods of exposure control indicate that many establishments which are estimated to incur some cost of compliance have overexposures in all processes at that establishment.

Survey respondents identified the presence of 10 different substances in SIC 40. Methyl alcohol was estimated to occur the most frequently at a total of 36 processes. Methyl alcohol was identified in maintenance activities. In addition, welding fumes occurred frequently in maintenance and welding activities.

SIC 42 - Motor Freight Transportation and Warehousing

Grain elevators are classified in three different two-digit SIC codes:

SIC 20, Food and Kindred Products; SIC 42, Motor Freight Transportation and Warehousing; and SIC 51, Wholesale Trade. Elevators falling within SIC 42 are those whose primary income derives from the storage of grain. Rulemaking participants who commented on the feasibility of achieving OSHA's proposed limit for grain dust in grain elevators did not designate SIC codes in their comments. The issue of grain dust exposure in grain elevators is discussed in connection with SIC 51, below. This SIC was not included in the 1988 survey.

SIC 45 - Air Transportation

This SIC includes establishments that provide domestic and foreign transportation by air and also those that operate airports and flying fields and provide terminal services. The Federal Aviation Administration (FAA), U.S. Department of Transportation, enforces rules and regulations governing the safety and health of flight and cabin crew of aircraft in flight. In general, the FAA also has jurisdiction over airline maintenance and ground/support personnel. According to the U.S. Department of Commerce, the U.S. airline industry consists of approximately 200 individual commercial air carriers operating over 4,400 aircraft and employing over 435,000 people [3]. In 1986, the industry served 418 million passengers and operated 7.4 billion freight and express cargo ton-miles. Nine major carriers account for 90 percent of all revenue passenger miles. (The U.S. Department of Commerce defines a major carrier as having annual revenues exceeding $1 billion, in 1982 dollars.) The remaining passenger revenue is shared by 16 carriers classified as nationals (each with annual revenues between $75 million and $1 billion in 1982 dollars), which account for about 12 percent, and by the regionals/commuters, which account for 4 percent. The U.S. Department of Commerce estimated the 1987 operating revenue for the airline industry as $55.6 billion, representing an annual growth rate of about 10 percent [3]. Revenue passenger miles were estimated as 435 billion, which represents a 3 percent rate of growth [4, p. 59-1]. In 1985, the median rate of return on assets in this sector was 4.3 percent [6].

In SIC 45, establishments generally employ dedicated maintenance workers to perform maintenance functions. The establishments in this SIC have up to five processes and as many as eight substances. Almost 60 percent of the processes involve potential exposure to these chemicals or substances on an intermittent short-term basis (up to 30 minutes). Fifty percent of the establishments in SIC 45 have reported the adoption of internal exposure standards. Among those establishments with internal standards, most use OSHA PELs.

The survey found that one third of the interviewed establishments had totally enclosed processes, but more than 65 percent were located outdoors. Of those establishments with ventilation systems, all were locally exhausted. More than 15 percent of the respondents reported having respirators for employee use. The combined data on exposure levels and methods of exposure control indicate that many establishments which are estimated to incur some cost of compliance have overexposures in all processes in that plant.

Survey respondents in SIC 45 identified 10 different substances. Ethylene glycol was estimated to occur the most frequently at 22 processes.

SIC 47 - Transportation Serices

SIC 47 includes establishments that furnish services related to transportation. Activities classified in SIC 47 include freight forwarding, arranging transportation for passengers and freight, renting railroad cars, inspection and weighing services; and freight car loading [1, pp. 280-281]. According to the U.S. Department of Commerce, between 1980 and 1985, SIC 47 was the third-fastest growing industry group, behind SIC 62 (Securities) and SIC 73 (Business Services) [3, pp. 13-14]. Between 1981 and 1985, SIC 47 experienced a 31 percent increase in employment. Table C-1 presents employment and establishment data for SIC 47. The median return on assets in this SIC was 7.1 percent [6].

In SIC 47, the survey identified more than eight times as many small firms (fewer than 20 production workers) as large firms. In the small firms maintenance work is predominantly performed by outside contractors. However in large firms a high percentage of maintenance work is performed by production workers.

The establishments classified in SIC 47 usually have one to four basic processes, with potential exposure to one to four substances. Ninety-four percent of the processes involve exposure to these chemicals or substances on an intermittent short-term basis (up to 30 minutes). Fifty percent of the firms in this SIC reported the adoption of an internal exposure standard. Of those large firms with exposure standards, most rely on OSHA PELs. Among small firms with exposure standards, most have adopted ACGIH TLVs. Internal monitoring had been performed at 50 percent of the processes.

The survey found that local exhaust ventilation is used most frequently to control exposures at processes not enclosed. Nearly 50 percent of the processes with chemical exposures have respirators for employee use, with a higher percentage of large firms reporting the presence of respirators than small firms. The combined data on exposure levels and methods of exposure control indicate that many establishments which are estimated to incur some cost of compliance have overexposures at all processes at that establishment.

Survey respondents identified the presence of 5 different substances in SIC 47. Exposure to gasoline was estimated to occur most frequently at a total of 206 processes. Gasoline was identified in loading/offloading/receiving/handling.

SIC 49 - Electric, Gas and Sanitary Services

SIC 49 includes establishments that generate, transmit, and/or distribute electricity, gas, or steam. These establishments may be combinations of any of these services, but also may include other types of services, such as transportation, communications, refrigeration and pipelines for natural gas. Water and irrigation systems, and sanitary systems that collect and dispose of garbage, sewage, and other wastes, also are included in this SIC [1, p. 284].

In recent years the utilities covered in SIC 49 have been affected by ongoing changes in regulations regarding utility rates and competition. Some industrial customers have begun producing their own energy and utilities are now competing for customers outside their service areas. This competition has forced structural change and diversification. Utilities have been forced to upgrade their overall efficiency. With declining interest rates, regulators have been decreasing the allowed rate of return for utilities. This, too, has led to intensified pressures on competition [21, p. 56]. The Federal Energy Regulatory Commission is currently considering whether to allow utilities to open their power lines to other competing utilities. Users would be given the choice of suppliers. With the decreasing rate of return and the increasing competition, utilities have stepped up efficiency in order to offset the impending drop in their profit margins [22, p. 48].

Many of the industries in SICs 4911, 4931, 4932, and 4939 are represented by the national trade association, Edison Electric Institute (EEI). Ninety-seven percent of all customers serviced by the investor-owned segment of the industry purchase electricity from EEI members. Members generate 76 percent of the country's electricity [Ex. 3-831].

Table C-1 presents employment and establishment data for SIC 49 for 1985. Between 1981 and 1985, SIC 49 experienced a 6 percent growth in employment. In 1985, almost 80 percent of all employees were production workers [2]. The median return on assets was 4.0 percent [6].

Within this SIC, there are seven three-digit SICs, including establishments that generate, transmit, or distribute electrical energy for sale and that operate crude petroleum and natural gas field properties; establishments that transmit and/or store natural gas for sale; establishments that provide electric or gas services in combination with other services, only if one service does not constitute 95 percent or more of revenues; establishments that distribute water for sale for domestic, commercial, and industrial use; establishments that collect and dispose of wastes conducted through a sewer system, including such treatment processes as may be provided; establishments that produce and/or distribute steam and heated or cooled air for sale; and establishments that operate water supply systems for the purpose of irrigation [1, pp. 284-286].

In SIC 49, the survey identified nearly twice as many small firms (fewer than 20 production workers) as large firms. In small firms, maintenance work is predominantly performed by production workers, although some firms also use dedicated maintenance workers or outside contractors. Large firms mainly employ workers specifically for maintenance duties.

The manufacturers classified in SIC 49 usually have one to four basic processes, with numerous firms having potential exposures to as many as six different substances. Fifty-five percent of the processes involve exposure to these chemicals or substances on an intermittent short-term basis (up to 30 minutes), with small firms tending to have more long-term exposures. Most firms in this SIC have adopted OSHA PELs as their internal standard. Of the large firms with internal exposure standards, most have adopted OSHA PELs or ACGIH TLVs. However, 52 percent of small firms reported having an internal exposure standard. Employee monitoring had been performed at 16 percent of the processes.

The survey found that about 25 percent of the processes are totally enclosed and 70 percent are located outdoors. Local exhaust and general ventilation are used frequently to control chemical exposures at processes. Nearly 45 percent of the firms with chemical exposures have respirators for employee use, with a higher percentage of large firms than small firms reporting the presence of respirators. The combined data on exposure levels and methods of exposure control indicate that many plants which are estimated to incur some cost of compliance do not have overexposures in all processes at that plant.

Survey respondents identified the presence of 47 different substances in SIC 49. Chlorine was estimated to occur most frequently at a total of 2,196 processes. Chlorine was identified in boilers, water treatment, handling spills/leaks, incineration, maintenance activities, use of chemical additives, use of disinfectants and solvents, and water purification.

SIC 50 and SIC 51 - Wholesale Trade

The wholesale trade sector includes establishments engaged in the wholesale selling of merchandise to retailers; industrial, commercial, institutional, farm, or business users, or to other wholesalers or firms that act as agents or brokers in the wholesale buying or selling of merchandise. Wholesale trade is divided into trade in durable goods (SIC 50) and in nondurable goods (SIC 51). This analysis focuses only on a few of the wholesale trade industries (e.g., dealers in scrap and waste materials, SIC 5093; grain, SIC 5191; and paints, varnishes, and supplies, SIC 5198 [1, pp. 24, 250, 255-257].

Wholesale trade sales ($1,375 billion in 1985) were fairly equally divided between durable goods and nondurable goods -- 46 percent and 54 percent, respectively [3, p. 56-1]. Of the approximately 425,000 establishments in wholesale trade, about five-eighths were in durable goods, and three-eighths were in nondurable goods. The specific four-digit industries studied for this analysis include about 11 percent of all wholesale trade establishments [2, pp. 59, 62, 64-65].

Table C-1 shows employment data at the four-digit level. Somewhat less than 60 percent of total employment in wholesale trade is in durable goods, while a little more than 40 percent is in nondurable goods. The specific four-digit industries being analyzed here account for less than 9 percent of all employment in wholesale trade [7].

OSHA received no comments relating to SIC 50, Wholesale Trade, Durable Goods. However, in SIC 51, Wholesale Trade, Non-Durable Goods, OSHA received several comments.

A large number of SIC 51 commenters submitted data and testimony directed at OSHA's proposed 4-mg/m(3) TWA exposure limit for grain dust [see, for example, Exs. 3-47, 3-58, 3-59, 3-65, 3-110, 3-281, 3-347, 3-387, 3-496, 3-667, and 3-752]. Several of these commenters provided estimates of the number of grain elevators potentially affected by the proposed limit, which pertains only to wheat, oat, and barley dusts. (Elevators used exclusively for storage are classified in SIC 4221; those used principally for grain cleaning and preparation fall within SIC 0723; and elevators that are used primarily in wholesale marketing operations are classified in SIC 5153, Wholesale Trade, Grain and Field Beans.) The National Grain and Feed Association (NGFA) estimates that approximately 87 percent of U.S. grain elevators handle oats, wheat, or barley, based on USDA records for 1985 that show that this percentage of government-grain-storing elevators reported handling and storing grains of these types [Ex. 3-752]; this yields an estimated total of 12,158 grain elevators that are potentially affected by the proposed grain dust limit. The NGFA estimates that approximately 49,063 full-time equivalent employees work in these elevators [Ex. 3-752].

According to the NGFA, many of the smaller grain elevators have relatively low profits:

A January 1987 survey and analysis by [the] U.S. Department of Agriculture on cooperative grain elevators (Financial Profile of Cooperatives Handling Grain: First Handlers, $1 Million to $4.9 Million in Sales. USDA, ACS Report No. 58, January 1987) indicates the average annual profits for small elevator facilities is only $38,272. This report indicated that 24.8 percent of these facilities currently have negative profits and another 23.1 percent have profits of only $25,000 or less [Ex. 3-752, pp. 19-20].

In these SICs, 19 out of 20 firms identified by the survey were small firms (fewer than 20 production workers). In the small firms, maintenance work is usually performed by production workers, but over 40 percent of all small firms use either outside contractors or have a dedicated maintenance staff. Large firms normally have a dedicated maintenance staff, but over one-quarter use production workers for maintenance and some use outside contractors.

Of those firms providing chemical or process information, the majority used from one to three chemicals in one or two processes. However, some firms reported using up to 10 chemicals. Over half of all exposures are short-term (up to 30 minutes), with the remaining exposures varying in length from 1 to 8 hours. The majority of small firms have internal exposure standards; most of these use OSHA PELs, but some use ACGIH TLVs. Almost 90 percent of large firms reported using internal exposure standards, and were about equally divided between those using OSHA PELs and those using ACGIH TLVs. Air monitoring was being done for approximately one-fifth of the processes found in large plants.

The survey found that less than 20 percent of the processes are totally enclosed and that over 50 percent of the processes are located outdoors. In almost 40 percent of the firms with chemical exposures, production workers use respirators, with large firms reporting a higher percentage of use than small firms.

Survey respondents identified the presence of 80 chemicals in SICs 50 and 51. Grain dust was estimated to occur the most frequently at a total of 3,426 processes, including drying, packaging/repackaging, receiving and shipping, sorting and grinding.

SIC 55 - Automotive Dealers and Service Stations

This industry sector includes retailers of transportation equipment for personal use (new and used automobiles) as well as recreational vehicles (boats, motor homes, and dune buggies); sellers of automobile parts and accessories; and gasoline stations. Although this industry does not include establishments whose primary business is automotive repair, it does include repair operations that are part of automobile dealerships or service stations. Only those retail outlets that earn more than 50 percent of their revenues from gasoline or lubrication oil sales are included. Many car washes and convenience stores that sell gasoline are excluded, as are traditional full-service gas stations that earn more than 50 percent of their revenues from such activities as repairs, towing, or the sale of auto accessories [1, pp. 265-266]. According to one estimate, this sector includes only 55 percent of all retail motor fuel outlets [23, pp. 6-13]. Although many employees are involved in selling, some are exposed to chemicals during painting or stripping or as a result of the indoor operation of engines or the use of solvents.

As shown in Table C-1, most establishments are relatively small (80 percent have fewer than 20 employees). Only in one sector, new and used automobile dealerships, do more than half of the establishments have more than 19 employees [10]. Even in this sector, however, 90 percent of the establishments have fewer than 100 employees [5]. Although the typical operation is relatively small, total employment is substantial because of the large number of establishments. New and used automobile dealerships account for 48 percent of total employment, gasoline service stations for 31 percent, and automobile and home supply stores for 16 percent.

Although many firms own only a single establishment, large firms own a significant portion of all establishments, which are operated as chains under leasing or franchising agreements.

The profitability of firms in SIC 55 is below the national average, with a median return on assets of 5.9 percent in 1986; however, this rate of return improved in 1986 as gasoline prices declined and new car sales increased [6].

In the survey, the analysis of SIC 55 was combined with SIC 75, automobile sales and service. The results for the two sectors combined will be reported here. In these two SICs, 19 out of 20 firms identified by the survey were small firms (fewer than 20 production workers). About two-thirds of the small firms use production workers to perform maintenance work. Over 70 percent of large firms have a dedicated maintenance staff; the remaining large firms are about equally divided between the use of outside contractors and the use of production employees for maintenance.

Of the firms reporting chemical or process use, most use from one to five chemicals in one to four processes, but 10 percent report using up to 10 chemicals. Employees are most commonly subject only to short-term exposures (up to 30 minutes) with less than one-fourth of firms reporting exposures of from 4 to 8 hours' duration. In these sectors, a majority of small firms have internal exposure standards. Those that use internal exposure standards are evenly divided between the use of OSHA PELs and the use of ACGIH TLVs. Almost 90 percent of large firms reported using internal exposure standards, usually OSHA PELs. Air monitoring data was being done for less than 10 percent of the processes found in large plants.

The survey found that approximately 10 percent of processes are totally enclosed and that less than 20 percent are located outdoors. In over 40 percent of firms with chemical exposures, production workers use respirators, with large firms having a higher percentage of use than small firms.

Survey respondents in SICs 55 and 75 identified the presence of 40 different substances. Carbon monoxide was estimated to occur the most frequently at a total of 109,093 processes, including cleaning, confined space exposure and maintenance activities. Gasoline was estimated to occur at 24,548 processes and toluene at 23,629.

SIC 72 - Personal Services and SIC 73 - Business Services

The personal services industry consists primarily of consumer services. SIC 721, laundry, cleaning and garment services has the highest potential for overexposure to chemicals. Other segments of SIC 72 include photographic studios (SIC 722); beauty shops, barber shops and shoe repair (SIC 723-725); and funeral service and crematories (SIC 726) [1, pp. 298-300].

As seen in Table C-1, the number of establishments in 1985 totaled 161,004. Almost all of these (96.9 percent) had fewer than 20 employees in 1985. The mean establishment size was 7 employees. The largest single segment of this industry is SIC 7231, beauty shops, which totaled 53,165 firms in 1986 [7]. Total employment (1,056,000 employees in 1985) has increased over the last several years. In 1986, the value of sales was $39.4 billion in the personal services industry, a 6.6 percent increase over 1985 [7]. The median rate of return on assets for the personal services industry was 10.5 percent in 1985 [6].

The dry cleaning industry is likely to be affected by the final rule's PEL for perchloroethylene. According to the 1982 Census of Service Industries, there were 13,049 dry cleaning plants in the U.S. that used perchloroethylene, with total employment of 89,896 workers.

The Amalgamated Clothing and Textile Workers Union (ACTWU) commented on the health risks associated with exposure to perchloroethylene. They also commented on the feasibility of reducing the exposure of perchloroethylene to substantially less than the 50 ppm proposed standard. (OSHA notes that, in the final rule, the PEL for perchloroethylene has been reduced to 25 ppm.) The International Fabricare Institute (IFI) supported the proposed revision of the PEL for perchloroethylene to 50 ppm. They stated that approximately 64 percent of the dry cleaning industry uses dry-to-dry equipment, and that over the past four years about 95 percent of all new equipment sold has been dry-to-dry equipment. The exposure to perchloroethylene is substantially greater when using transfer equipment versus dry-to-dry equipment [Ex. 8-31]. In 1982 there were 6,738 dry-to-dry machines compared to 12,929 dry-to-dry machines in 1988 [Ex. 3-671].

The business services industry (SIC 73) consists of several different sectors. Among the sectors included are mailing, reproduction, and commercial art and photography (SIC 733); building cleaning and maintenance services (SIC 734); and miscellaneous business services (SIC 739), such as photofinishing laboratories and commercial testing laboratories [1, pp. 301-308].

The number of establishments in the business services industry in 1985 totaled 382,626. Almost all of these (90.5 percent) had fewer than 20 employees in 1985. The mean establishment size was 12 workers. Total employment (4,457,000 employees in 1985) has increased over the last several years (4,057,000 employees in 1984) [4]. In 1986, the value of sales was $198.7 billion in the business services industry, a 9.2 percent increase over 1985 [7]. The median rate of return on assets in SIC 73 was 11.1 percent in 1985 [6].

In these SICs, the survey identified over 20 times as many small firms (fewer than 20 production workers) as large firms. In the small firms, maintenance work is performed for the most part by production workers, although some firms use outside contractors and some firms employ maintenance staffs. Large firms divide maintenance work about equally between a dedicated maintenance staff, production workers, and outside contractors.

Employees are exposed to chemicals on an intermittent short term basis (up to 30 minutes) or continuously (up to 8 hours per day), with small firms tending to have more short-term and long-term exposures. Small firms generally have no internal exposure standards; when they do, the OSHA PELs are followed about one-half of the time. Over three-fourths of large firms reported using the OSHA PELs; the balance indicated that they have no standards or they rely on ACGIH TLVs. Air monitoring was being conducted at about one-tenth of the processes found in large plants.

The survey found that about one-fourth of the processes are totally enclosed and less than 1 percent are located outdoors. Local exhaust ventilation and general dilution are used most frequently to control exposures at processes not enclosed or outdoors. In one-tenth of the firms with chemical exposures, production workers use respirators, with a higher percentage of large firms using respirators than small firms. The combined data on exposure levels and methods of exposure control indicate that overexposure is not occurring at any processes in small firms and at less than one percent of the processes in large firms.

Survey respondents in this SIC identified eight major processes:

permanents, dry cleaning, manicure/pedicure, coloring/dyeing, embalming, washing, exterminating, and photofinishing. Chemicals that were present in these processes included: acetic acid, acetone, calcium hydroxide, chlorine, chlorpyrifos, diazinon, ethylene glycol, glutaraldehyde, iron oxide, isopropyl alcohol, methyl alcohol, methyl chloroform, perchloroethylene, naphtha, sodium hydroxide, toluene, trichloroethylene, xylene, and titanium dioxide.

SIC 75 - Automotive Repair, Services, and Garages

This sector includes establishments that provide automotive repair, rental, leasing, and parking services to the general public, but excludes gasoline stations (SIC 55) and repair shops that are part of automobile dealerships or that service commercial fleets [1, p. 309]. Employees may be exposed to engine emissions in parking garages or repair shops, to a variety of chemical solvents (particularly in painting and stripping), and to dust from body work.

Eighty-five percent of the establishments are automotive repair shops, which is the sector most likely to have significant chemical exposure, and they employ 61 percent of all industry workers [10]. As shown in Table C-1, SIC 75 is dominated by businesses employing fewer than 20 workers (97 percent), with a median return on assets of 9.2 percent in 1985. The profitability of automotive repair and service firms is high, although it varies by size and industry sector. Small firms (under $100,000 in assets) had returns of 18.3 percent in 1985, while large businesses (over $1,000,000) had returns of 3.9 percent. Paint shops (SIC 7535) were the most profitable type of operation, while parking lots (SIC 7523) and parking structures (SIC 7525) registered significantly lower rates of return [6].

No rulemaking participants submitted comments on this sector, and the results of the survey are reported under SIC 55.

SIC 76 - Miscellaneous Repair

This industry group includes a wide variety of repair services, differentiated by object repaired and processes used. Industries of particular concern include reupholstery and furniture repair (SIC 7641) and welding (SIC 7692) [1, pp. 312-314]. Reupholstery and furniture repair workers may be exposed to wood dust during wood working, and to solvents; welders may be exposed to fumes.

Nineteen percent of the 56,000 industry establishments in SIC 76 are in SIC 7641 and SIC 7692. These two industries account for approximately 14 percent of all SIC 76 employment [7, pp. 81-82].

The industry is made up almost entirely of very small firms, and the sector has extremely low concentration. Mean business size is 5.5 employees: more than 95 percent of all establishments have fewer than 20 employees, and 65 percent of all workers are employed by establishments of this size. Only 0.2 percent of all miscellaneous repair establishments (with about 6 percent of total employment) have 100 or more employees, and only 17 establishments have 250 or more. The four-digit industries of concern are even more completely dominated by small establishments, with a mean size of 4.8 employees in SIC 7641 and 3.4 employees in SIC 7692 [7, pp. 81-82].

Despite a slight decline in 1981 and 1982, employment in SIC 76 has grown fairly steadily, increasing by 23 percent between 1979 and 1984 and by 7 percent between 1984 and 1986.

Miscellaneous repair firms have high profit rates. The median 1985 rate of return on assets in SIC 76 was 10.0 percent. This rate of return was higher than that of any two-digit manufacturing industry. The median rates of return on assets in SIC 7641 and SIC 7692 are over 11 percent [6].

In this SIC, over 99 of 100 firms identified by the survey were small firms (fewer than 20 production workers). In the small firms, maintenance work is normally performed by production workers, with some use of dedicated maintenance workers or use of outside contractors. In the large firms, maintenance work is performed by either dedicated maintenance staff, production employees, or outside contractors, with a dedicated maintenance staff being the most common.

Most firms reporting process or chemical use report using one to four chemicals in one to three processes. For small firms, short-term exposures (up to 30 minutes) are the most common, with the remaining exposures ranging from 1 to 8 hours in length. In large firms, most exposures are for longer than one hour. Over three-quarters of large firms in this sector report using internal exposure standards; over 55 percent reported using OSHA PELs. A majority of small firms reported having internal exposure standards; more use OSHA PELs than use ACGIH TLVs. Air monitoring data had been collected for 26 percent of the processes found in large plants.

The survey found that just over one-quarter of the processes are totally enclosed and that one-quarter are located outdoors. In over 50 percent of the firms with chemical exposures, production workers use respirators; large firms use respirators more commonly than do small firms.

Welding fumes were identified most frequently by respondents in SIC 76 in some 4,040 welding and brazing processes.

No rulemaking participants submitted comments on this sector.

SIC 80 - Health Services

The health services industry encompasses a broad range of medical, surgical, and other health services, both public and commercially owned. these services are provided by a variety of practitioners (e.g., physicians, dentists, osteopathic physicians, chiropractors, optometrists) at a variety of facilities (e.g., hospitals, nursing facilities, outpatient care facilities, medical laboratories) [1, pp. 321-323].

Total expenditures on health care and medical services ($425 billion in 1985) are very large, with 40 percent of this amount going to hospital care and 20 percent to physicians' services. Expenditures on nursing home care, drugs and medical sundries, and dentists' services each accounted for 6 to 8 percent of all health and medical services expenditures [3, p. 54-1].

Data on health care establishments are shown in Table C-1. Although the number of health service establishments (313,000) is very large, 85 percent of these are offices of licensed practitioners. No other three-digit sector within the health services industry accounts for more than 4 percent of health service establishments.

Total health services employment is very large (6.3 million), with hospitals accounting for almost half (i.e., 48 percent) of this workforce. Because of their large numbers, practitioners' offices are next in percentage of workforce employed (24 percent), followed by nursing and personal care homes (18 percent). Mean establishment sizes range from six or fewer employees in practitioners' offices to 250 or more employees in hospitals. The overall mean size of establishments in this industry is 20 employees, with more than 91 percent of these establishments having fewer than 20 employees, and approximately 22 percent of all SIC 80 employees working in establishments of this size. SIC 80 facilities with more than 250 employees employ more than 50 percent of the workforce in this sector [7; 5].

The health and medical services industry has been expanding rapidly for more than a decade. A variety of factors have caused this increase, including the expansion of the elderly population and improved treatment for many illnesses. In addition, between 1985 and 1986, the price for most medical services rose between 6 and 9 percent, compared with 1.5 percent increase in consumer prices. The implementation of Medicare's prospective payment system is also causing major changes in the health care industry [3, pp. 54-1,2].

Hospital care costs have been a major target of cost-cutting measures, resulting in a decline in hospital admissions, a shortening of hospital stays, and substantial industry restructuring, including increased mergers and acquisitions by large chains, vertical integration, diversification of services offered, expanded professional peer review, and more businesslike operations. Major investor-owned nursing home chains also have experienced rapid expansion and acquisition [3, pp. 54-1, 2].

For SIC 80 as a whole, the growth rate in expenditures averaged 12.6 percent per year from 1979 to 1984 and more than 9 percent for the next 3 years [3, p. 54-1]. Employment grew by 31 percent between 1979 and 1986, rising by 2 to 5 percent in each year [5]. The growth picture is fairly consistent across three-digit industries, although expenditures on "other professional services" have shown the most rapid growth of any health service (16.3 percent annually from 1979 to 1984). Expansion has been especially rapid in health maintenance organizations and home health care, both of which have the potential for reducing health costs and substituting, to some degree, for hospital care [3, pp. 54-1 to 54-4].

The median rate of return on assets in health services (5.0 percent in 1985) is relatively low compared with that in manufacturing industries, and hospitals have somewhat lower median rates of return than is the case for health services as a whole. Several "offices" industries, on the other hand, have median rates of return higher than 13 percent. Medical and dental laboratories have median rates of return that are above the median for two-digit manufacturing industries [6].

In this SIC, nine out of ten firms identified by the survey were small firms (fewer than 20 production workers). The majority of small firms employ outside contractors for maintenance work. The majority of large firms use dedicated maintenance workers.

Of those firms reporting chemical or process use, over half report using one or two chemicals in one or two processes. In this SIC, most employee exposures are for less than 30 minutes in length; one-fourth of large firms report employee exposures of from 4 to 8 hours in duration. Approximately half of all small firms have no internal exposure standards and approximately half report using OSHA PELs. Over 60 percent of large firms report using OSHA PELs, with only a limited number reporting use of ACGIH TLVs. Large firms provided air monitoring data for slightly over one-third of all processes found in their plants.

The survey found that almost 50 percent of all processes are totally enclosed and that almost none are outdoors. One-quarter of all firms report using respirators, with large firms having somewhat greater respirator use than small firms.

Survey respondent identified the presence of 73 different substances in SIC 80. Isopropyl alcohol was estimated to occur the most frequently at a total of 38,575 processes, including administration of anesthesia, laboratory procedures, making of dental appliances and sterilization. Aryl and inorganic compounds of mercury were estimated to occur in a total of 25,197 processes, including preparation of dental amalgams and X-ray film processing.

No testimony or comments submitted to the docket pertained to the health industry.


                             TABLE C - 1
                Industries With Potential Hazardous Exposures;
                   Number of Establishments and Employment
                                (1985)
________________________________________________________________________
                                  Establishments(a)       Employment(d)
                               Total     Percent         Tot.  Production
SIC Description                Number Large(b) Small(c) (1,000)   Workers
                                                                  (1,000)
_________________________________________________________________________
20   FOOD AND KINDRED PRODUCTS   29,043  37.14  62.86     1,603     1,118
21   TOBACCO MANUFACTURES           216  46.76  53.24        64        48
22   TEXTILE MILL PRODUCTS       11,023  39.40  60.60       702       607
23   APPAREL PRODUCTS            30,032  33.33  66.67     1,121       945
24   LUMBER & WOOD PRODUCTS,
       EXCEPT FURNITURE          36,710  19.73  80.27       697       584
242  SAWMILLS, PLANING MILLS      6,390  68.95  31.05       195       172
243  MILLWORK, VENEER & PLYWOOD  13,921  17.87  82.13       288       190
244  WOOD CONTAINERS              2,701  78.08  21.92        41        35
245  BUILDING & MOBILE HOMES      1,618  40.05  59.95        72        56
249  MISCELLANEOUS WOOD PRODUCTS  5,666  18.73  81.27        77        64
25   FURNITURE AND FIXTURES      16,791  27.16  72.84       494       394
26   PAPER AND ALLIED PRODUCTS    8,750  53.86  46.14       678       512
27   PRINTING, PUBLISHING &        ALLIED INDUSTRIES         84,279  15.87  84.13     1,428       789
28   CHEMICAL AND ALLIED
       PRODUCTS                  20,823  32.59  67.41     1,044       578
281  INDUSTRIAL INORGANIC
       CHEMICALS                  3,024  35.42  64.58       142        72
282  PLASTICS & SYNTHETICS        1,666  51.50  48.50       172       114
283  DRUGS                        2,454  37.82  62.18       206        95
284  SOAP, CLEANERS, & COSMETICS  4,498  24.59  75.41       148        94
285  PAINTS, VARNISHES, LACQUERS  1,880  36.54  63.46        64        31
286  INDUSTRIAL ORGANIC CHEMICALS 1,528  34.88  65.12       160        82
287  AGRICULTURAL CHEMICALS       1,843  23.77  76.23        59        37
289  MISCELLANEOUS CHEMICAL       3,930  29.64  70.36        94        54
       PRODUCTS
29   PETROLEUM REFINING & RELATED
       INDUSTRIES                 3,334  28.40  71.60       179       109
291  PETROLEUM REFINING           1,332  33.18  66.82       141        82
295  PAVING & ROOFING MATERIALS   1,222  23.81  76.19        26        20
299  MISCELLANEOUS PETROLEUM &        COAL PRODUCTS                780  27.44  72.56         -         -
30   RUBBER & PLASTICS PRODUCTS  18,002  38.85  61.15       786       607
307  MISCELLANEOUS PLASTIC
       PRODUCTS                  14,638  39.62  60.38       550       435
31   LEATHER AND LEATHER PRODUCTS 3,940  29.85  70.15       165       137
311  LEATHER TANNING & FINISHING    480  35.42  64.58        15        12
32   STONE, CLAY, GLASS, &      CONCRETE PRODUCTS           21,054  26.26  73.74       588       451
33   PRIMARY METAL INDUSTRIES    10,101  44.75  55.25       808       612
34   FABRICATED METAL PRODUCTS   46,322  32.96  67.04     1,465     1,084
35   MACHINERY, EXCEPT
       ELECTRICAL                77,748  22.90  77.10     2,174     1,307
36   ELECTRICAL & ELECTRONIC
       MACHINERY, EQUIPMENT &        SUPPLIES                  28,478  37.64  62.36     2,197     1,300
37   TRANSPORTATION EQUIPMENT    16,132  31.58  68.42     1,980     1,257
38   INSTRUMENTS                 16,814  29.42  70.58       720       391
39   MISCELLANEOUS MANUFACTURING
       INDUSTRIES                32,212  15.82  84.18       367       264
40   RAILROAD TRANSPORTATION      2,645  27.30  72.70       359         -
45   TRANSPORTATION BY AIR       11,832  19.46  80.54       522         -
47   TRANSPORTATION SERVICES     35,626   7.56  92.44       276         -
49   ELECTRICAL GAS, & SANITARY
       SERVICES                  21,115  25.71  74.29       915       729
5093 SCRAP & WASTE MATERIALS      7,556  12.61  87.39        92         -
5153 GRAIN                        7,523   5.84  94.16         -         -
5161 CHEMICALS & ALLIED
       PRODUCTS                  13,045   8.51  91.49         -         -
5191 FARM SUPPLIES               20,392   4.55  95.45       151         -
5198 PAINTS, VARNISHES,           4,033   6.89  93.11         -         -
       SUPPLIES
55   AUTO DEALERS, SERVICE
       STATIONS                 189,214   9.77  90.23     1,890     1,886
72   PERSONAL SERVICES          161,004   3.13  96.87     1,056         -
73   BUSINESS SERVICES          382,626   9.46  90.54     4,457     3,863
75   AUTO REPAIR, SERVICES,
       & GARAGES                149,260   2.64  97.36       731       614
7641 REUPHOLSTERY, FURNITURE
       REPAIR                    10,655   0.92  99.08         -         -
7692 WELDING REPAIR               9,413   2.21  97.79         -         -
80   HEALTH SERVICES            313,076   8.71  91.29     6,299     5,607
_________________________________________________________________________
Source: U. S. Department of Labor Occupational Safety and Health
Administration, Office of Regulatory Analysis.

  Footnote(a) Dun and Bradstreet
  Footnote(b) 20 or more employees
  Footnote(c) Fewer than 20 employees
  Footnote(d) Labstat, U.S. Department of Labor (Database)

References

  1. Executive Office of the President. Office of Management and Budget.
Standard Industrial Classification Manual.  Washington, D.C.:  Government
Printing Office, 1972.
  2. U. S. Department of Commerce, Bureau of the Census, 1985 Annual
Survey of Manufactures: Industry Statistics, Washington, D.C., 1987.
  3. U.S. Department of Commerce, International Trade Administration, 1987
U.S. Industrial Outlook. Washington, D.C.: Government Prinint Office,
January 1987.
  4. U.S. Department of Commerce.  International Trade Administration,
1988 U.S. Industrial Outlook, Washington, D.C.: Government Printing
Office, January 1988.
  5. U.S. Department of Labor, Bureau of Labor Statistics.  LABSTAT data
base. 1987. (Unpublished Data.)
  6. Dun and Bradstreet, Inc. Industry Norms and Key Business Ratios,
1987. (Database)
  7. U.S. Department of Commerce.  Bureau of the Census.  Country Business
Patterns 1984 and 1985. Washington, D.C.: 1986. (Database)
  8. Telephone communication with Donald W. Butts and Barbara Wise, U.S.
Department of Commerce.  Office of Forest Products and Domestic
Construction. January 19, 1988.
  9. U.S. Department of Commerce.  Bureau of the Census. 1985 and 1986
Annual Survey of Service Industries: Industry Statistics.  Washington,
D.C., 1987. (Database)
  10. Dun and Bradstreet, Inc. Dun's Market Identifiers. 1987. (Database)
  11. Telephone conversation with Charles Bell, U.S. Department of
Commerce, December 23, 1988.
  12. Wall Street Journal, December 27, 1988, p C12.
  12. Value Line, Inc. Industry Surveys, August 5, 1988, p. 1077.
  13. Value Line, Inc. Industry Surveys, August 12, 1988, p 1221.
  14. Telephone conversation with David Larabee, U.S. Department of
Commerce, December 23, 1988.
  15. Wall Street Journal, December 20, 1988, p C14.
  16. Telephone conversation with David Climert, U.S. Department of
Commerce, December 23, 1988.
  17. Value Line, Inc. Industry Surveys, July 19, 1988, p. 1349.
  18. Wall Street Journal, December 27, 1988, p. A2.
  19. Value Line, Inc. Industry Surveys, August 19, 1988, p. 1337.
  20. Value Line, Inc. Industry Surveys, August 5, 1988, p. 1076.
  21. Value Line, Inc. Industry Surveys, July 19, 1988, p. 603.
  22. Utilities industry earnings growth by company projection, 1987,
Financial World, January 7, 1987.
  23. Sobotka and Co., Inc., Regulatory Impact Analysis for Technical
Standards for Underground Storage Tanks. Washington, D.C.: 1987.

D. Employee Exposures And Benefits

Employee exposures to the substances included in the scope of this rulemaking are associated with a wide variety of acute and chronic conditions and illnesses. These include sensory irritation, narcosis, organ system dysfunction, chronic respiratory disease, neurological impairment, allergic sensitization, and cancer. Since OSHA's adoption of existing Federal and consensus standard limits in 1971, toxicologic evidence has become available that shows that adverse health effects can occur as a consequence of exposure to many of the substances listed in OSHA's Z tables, and that such health effects occur even when exposures are maintained at the current Z-table limits. In addition, many substances that have come into widespread use or been introduced since 1971 have been shown to be potentially hazardous in the workplace environment. OSHA thus believes that reducing worker exposures to such substances by lowering existing exposure limits or by adding limits for previously unregulated substances will result in a significantly reduced risk of illness to workers.

This chapter describes both the methodology used to identify workers potentially exposed to the hazardous substances included in this rulemaking and the expected benefits to those workers resulting from lowering permissible exposure levels. An important existing data base for identifying employees potentially exposed to hazardous substances was OSHA's Integrated Management Information System (IMIS). The IMIS data were used to project expected benefits resulting from lowering permissible exposure levels of the substances being regulated.

The IMIS data base, however, was incomplete, and its information on some hazardous chemicals may be out of date. For example, IMIS contained research information on about 160 substances among the approximately 430 substances covered by the final rule. While the IMIS data base contains the results for over 100,000 samples of substances currently regulated by OSHA, no plant-specific information was available for about 200 of the substances included in this rulemaking but currently not being regulated by OSHA. To both correct this data gap and obtain additional information on employee exposures, a nationwide survey was begun in January 1988, which was designed to collect worker exposure data from about 5,700 establishments nationwide in industries that are believed to be affected by this rulemaking. The survey results include industry-sector-specific data on the extent of employee exposures to hazardous materials and, unlike the IMIS data, provide specific information on the industrial processes in which these substances are used. While the sample survey confirmed potential exposures for many of the 160 chemicals in the IMIS data base, it identified potential exposure problems for about 62 additional substances subject to this rulemaking. Thus, the benefit estimates in this section are based upon employee exposures to 212 of the 428 substances being regulated.

To assess the benefits of revising OSHA's Z tables, OSHA relied on both the survey and IMIS data. The IMIS data were combined with raw survey data to estimate the extent to which employees are currently exposed to substances included in this rulemaking. From this analysis, OSHA estimated the reduction in illness cases and disease-related fatalities associated with reducing exposure limits for these substances.

Description of Data Sources Used

To assess the quantitative benefits associated with this rulemaking, the following data were used:

* The exposure limits for substances included in the rulemaking;
* Employee exposure data for these substances;
* Employment data by four-digit SIC code for the base year 1985;
* Annual illness and lost workday rates for the base year 1985; and

* Health effects information on the substances included in the rulemaking.

Employee exposure data for about 160 substances were obtained from OSHA's Integrated Management Information System (IMIS). This data base contains exposure measurements obtained by OSHA compliance officers during the conduct of thousands of health inspections. For each facility inspected, the IMIS file includes information on the number of employees at the facility, results of employee air monitoring for specific substances, and the number of employees potentially exposed to each substance monitored. To perform the benefits assessment, a summary IMIS file was created that contained the following information:

* A list of substances for which personal 8-hour TWA samples were taken, by four-digit SIC and facility inspected;

* The number of workers potentially exposed to each substance monitored, by four-digit SIC and facility;

* The number of employees at each facility inspected;

* The total number of personal 8-hour TWA samples obtained for each substance, by four-digit SIC and facility; and

* The number of samples taken at each facility that showed concentrations exceeding OSHA's limits.

Only those substances for which OSHA is reducing an existing 8-hour TWA limit or adding a new 8-hour TWA limit were included in the analysis. A total of approximately 37,500 personal air sample results for about 160 substances were appropriate for use in this analysis. This analysis does not estimate the benefits associated with reducing current ceiling limits or adding new short-term exposure limits (STELs), either because the data obtained from the IMIS did not include information on sample duration for ceiling or peak measurements, or because OSHA was not able to relate the IMIS data on ceiling or peak measurements to the final short-term or ceiling limits.

In addition to relying on the IMIS exposure data, OSHA completed a major telephone interview survey of about 5,700 workplaces that are potentially affected by the revision of OSHA's Z tables. Data from this survey provide information on substances that are used in a variety of industrial processes at the facilities surveyed, on the number of workers involved in those processes, and on whether personal exposure measurements taken at the processes exceeded OSHA, ACGIH, or NIOSH limits.

Employment data by four-digit SIC code were obtained from three data sources. For each four-digit SIC represented in the IMIS file, OSHA first relied on 1985 data from the BLS LABSTAT data base [1]. Where data were unavailable from this source at the four-digit SIC level, OSHA relied on Dun & Bradstreet's Market Identifiers file for 1985 [2]. Data from 1985 County Business Patterns [3] were used to obtain employment data for four-digit SIC groups not represented in either the LABSTAT or the Dun & Bradstreet file.

Data on illness and lost workday rates were obtained from the 1985 LABSTAT file for all industries (at the three- and four-digit level) represented in the IMIS file. These data included rates per 100 employees for total illness cases, lost-workday illness cases, and total number of lost workdays.

Estimates of the Number of Potentially Exposed Employees

Estimates of the number of employees potentially exposed to the substances included in this analysis were derived from the IMIS data, OSHA's survey data, and employment data bases. To conduct the analysis, OSHA used the IMIS and survey data separately to derive independent estimates of the number of workers potentially exposed and the number of workers exposed above the limits for each substance. The estimates derived from these two data sources were then combined to yield an overall assessment of the extent of employee exposure, by four digit SIC, to substances included in this rulemaking. The following sections describe how each of the data bases was used to develop estimates of employee exposures, and how these estimates were then combined.

Estimates Derived From OSHA's IMIS Data Base. For each facility inspected, the IMIS contained information on the number of employees at the facility and the number of employees observed to be potentially exposed to each substance for which personal air samples were collected. For each substance sampled within an industry (at the four-digit level), the estimated number of employees potentially exposed to that substance in the industry was determined by the following formula:

                Sigma of    P(f) * W = P
                           _____
                Sigma of    E(f)

where
  P(f) = number of employees observed to be potentially expose to the
substance at a facility;
  E(f) = total number of employees at the facility;
  W    = number of production workers in the industry in 1985; and
  P    = estimated number of employees potentially exposed to the substance
in the industry.
  The estimated number of workers currently exposed above the limits for
each substance was calculated using the following formula:

                 Sigma of  S(f) * P = Z
                           _____
                 Sigma of  T(f)
where
  S(f) = number of samples that exceeded the limit for the substance at
all facilities in an industry sector;
  T(f) = total number of personal samples taken for the substance at
all facilities in the industry sector;
  P   = estimated number of employees potentially exposed to the substance
in the industry sector; and
  Z   = estimated number of workers in the industry sector currently
exposed above the limits for the substance.
  Estimates Derived From OSHA's Survey Data.  Facilities participating in
OSHA's telephone survey provided the following information that was useful
for estimating the extent of employee exposures to chemical substances:
  * The facility's four-digit SIC code;
  * The total number of production employees at the facility; at the
facility;
  * The substances used or present in each process;
  * The exposure limits used as internal targets or goals at the facility
(i.e., OSHA's current limits, ACGIH limits, NIOSH limits, or "Other"
limits such as those from material safety data sheets or insurance
carriers); and
  * Whether employee exposures exceeded the targeted limits for each
process/chemical combination present at the facility.
  To estimate the number of employees potentially exposed to a given
substance in a four-digit SIC industry group, OSHA assumed that all
employees who are involved with processes in which the substance is used
or present are potentially exposed.  Thus, the formula for estimating the
number of employees who are potentially exposed to a substance in a given
industry sector is

                Sigma of  X(f) * W = P
                          ____
                Sigma of  T(f)
where
  X(f) = number of employees at the facility who are involved in
processes using a given substance;
  T(f) = total production workforce at the facility;
  W    = the number of production workers in the industry sector in 1985;
and
  P = estimated number of employees potentially exposed to the substance
in the industry sector.

To estimate the number of employees currently exposed above the final limits, OSHA relied on survey responses that indicated whether exposure measurements associated with a process exceeded the facility's internal exposure limits. OSHA interpreted the survey responses as follows.

* OSHA assumed that none of the potentially exposed employees are currently exposed above the limit for any substance associated with a process if (1) The revised limit is an ACGIH TLV and respondents indicated that exposure measurements did not exceed ACGIH, NIOSH, or some "other" set of limits, or (2) The revised limit is a NIOSH REL and respondents indicated that exposure measurements did not exceed NIOSH limits.

* OSHA assumed that all of the potentially exposed employees are currently exposed above the limits for all substances associated with a process if (1) the revised limit is an ACGIH TLV and respondents indicated that exposure measurements did exceed OSHA, ACGIH, or some "other" set of limits, or (2) the revised limit is a NIOSH REL and respondents indicated that exposure measurements did exceed OSHA, ACGIH, NIOSH, or some "other" set of limits.

The number of overexposed workers were then summed for each substance across all facilities that responded to the survey in the four-digit SIC industry group. In instances where the survey data yielded no information on whether employees are or are not exposed above the final rule limit for a substance, and no exposure data were available from IMIS on that substance, the number of workers exposed above the final rule limit for that substance is unknown (this is indicated in Supplement 2 by a blank space). Since it is likely that in some of these cases there are employees exposed above the final rule limits, OSHA believes that it has not necessarily accounted for all employees who are exposed above the final rule limits for the 212 substances included in this analysis. Thus, OSHA believes that the number of overexposed employees may be understated.

Since the publication of the PRIA, OSHA has identified the chemical composition of several generic and trade-name substances noted as being in use by survey respondents. The estimates of potential benefits presented in this final RIA include employees exposures to substances contained in these generic and trade-name products.

Approach for Combining Estimates Derived from the IMIS Data and Survey Data. To obtain an overall estimate of the extent of employee exposures to substances used in each four-digit SIC industry group, OSHA combined the estimates derived separately from the IMIS and survey data. Table D-1 illustrates how these estimates were combined to yield an overall estimate of the extent of employee exposures in SIC 2851. Where estimates for a given substance could be derived from one data set but not the other, the combined assessment uses the available estimates without adjustment. Where estimates could be derived from both data sets for the same substance, the combined assessment is based on the average of the available estimates; this approach has the effect of giving equal weight to estimates derived from either the IMIS or survey data.


Table D-1. Analysis of Employee Exposures in SIC 2851 Derived From IMIS
           Data, Survey Data, and Both IMIS and Survey Data Combined
--------------------------------------------------------------------------
SIC 2851
Paints and Allied Products

                  Assmnt. from IMIS   Assmnt. from Survey Combined Assmnt.
                  _________________   ___________________ ________________
                    Workers Workers   Workers   Workers   Workers  Workers
                Potentially   Above   Potent.     Above   Potent.    Above
Name                Exposed  Limits   Exposed    Limits   Exposed   Limits
__________________________________________________________________________

ACETONE               2,875       0    63,553         0    33,214        0
ALPHA-ALUMINA         1,286       0                         1,286        0
AMMONIA                                14,041         0    14,041        0
BUTOXYETHANOL         2,121       0                         2,121        0
N-BUTYL ACETATE                        70,204         0    70,204        0
BUTYL ACRYLATE       13,302       0    61,336         0    37,319        0
N-BUTYL ALCOHOL                        44,339    14,189    44,339   14,189
N-BUTYL GLYCIDYL
  ETHER              28,030       0                        28,030        0
CARBON MONOXIDE         672     672                           672      672
CARBON TETRACHLORIDE  5,912       0                         5,912        0
COBALT AS CO          4,604       0                         4,604        0
CYCLOHEXANONE         6,821       0                         6,821        0
DIISOBUTYL KETONE     4,434       0                         4,434        0
ETHYL ACRYLATE                         73,899         0    73,899        0
ETHYLENE GLYCOL                        46,556    17,691    46,556   17,691
ETHYLENE GLYCOL
  DINITRATE                            70,204         0    70,204        0
FURFURAL              2,956       0                         2,956        0
HEPTANE               5,454       0                         5,454        0
HEXAFLUOROACETONE    41,383  41,383                        41,383   41,383
HEXANE                4,678       0                         4,678        0
2-HEXANONE            6,547     727    52,468    28,858    29,508   14,792
HEXONE                7,131     319                         7,131      319
ISOBUTYL ALCOHOL     10,996       0                        10,996        0
ISOPHORONE                             42,122    16,007    42,122   16,007
ISOPROPYL ALCOHOL                      28,082         0    28,082        0
KAOLIN, TOTAL DUST                     69,465         0    69,465        0
MAGNESIUM OXIDE
  FUME, AS MG        10,560       0                        10,560        0
METHYL CHLOROFORM
  (1,1,1-
  TRICHLOROETHANE)                     73,899         0    73,899        0
METHYL ETHYL KETONE
  PEROXIDE                             25,126         0    25,126        0
MOLYBDENUM, INSOLUBLE
  COMPOUNDS AS MO    11,853       0                        11,853        0
PERCHLOROETHYLENE     1,973       0                         1,973        0
PETROLEUM DISTILLATES,
  RUBBER SOLVENT      7,885       0    68,726         0    38,306        0
PHTHALIC ANHYDRIDE    4,427       0                         4,427        0
PROPYL ALCOHOL                         26,604         0    26,604        0
SODIUM HYDROXIDE                       29,560         0    29,560        0
STODDARD SOLVENT      6,961     194    42,861    18,430    24,911    9,312
STYRENE               1,508       0    31,038    13,967    16,273    6,983
TALC (NON-ASBESTIFORM)                  5,912         0     5,912        0
TIN METAL AND OXIDE   1,286       0                         1,286        0
TITANIUM DIOXIDE      2,668       0    29,560    29,560    16,114   14,780
TOLUENE               7,538     187    25,865         0    16,701       94
TOLUENE-2,
  4-DIISOCYANATE                       17,736         0    17,736        0
TRIBUTYL PHOSPHATE      244       0                           244        0
TRICHLOROETHYLENE     3,695       0                         3,695        0
TRIETHYLAMINE           244       0                           244        0
TRIMELLITIC ANHYDRIDE 1,626     813    28,082         0    14,854      407
TRIMETHYL BENZENE     8,099       0                         8,099        0
VINYL ACETATE        13,302       0    30,299         0    21,800        0
VM & P NAPHTHA       25,909       0                        25,909        0
WOOD DUST            88,679  73,896                        88,679   73,896
XYLENE (O,M,P-ISOMERS)                 20,692         0    20,692        0
ZINC OXIDE (FUME)                      57,641         0    57,641        0
__________________________________________________________________________

Estimates of both the number of employees potentially exposed and the number exposed above the limits are presented, by substance and four-digit SIC code, in Supplement 2. This supplement also identifies, by four-digit SIC code, substances that are judged to present potential exposure problems but for which no IMIS or survey data were available.

Aggregate estimates of the number of employees potentially exposed or exposed above the final limits to any substance considered in the analysis are presented by two-digit SIC code in Table D-2. Because an employee may be exposed to more than one substance in a given industry, aggregate estimates of the size of the exposed population are presented as minimum and maximum estimates. Maximum estimates of the size of the exposed population assume that no employee is exposed to more than one substance; minimum estimates assume the greatest possible extent of multiple-chemical exposure. For example, if 200 employees are estimated to be exposed to acetone and 300 employees are estimated to be exposed to toluene in a given industry, a minimum of 300 employees is estimated to be exposed to both substances in the industry, and a maximum of 500 employees is estimated to be exposed to either substance in the industry.


Table D-2.  Number of Workers Exposed and Number of Workers for Whom Risk
            is Reduced, by 2-Digit SIC

Note: Because of its width, this Table has been divided; see continuation
for additional columns.
______________________________________________________________________
                                              # Workers    # Workers
                    # Workers    # Workers      Exposed      Exposed
                    Potent'ly    Potent'ly        Above        Above
        Number of    Exposed,     Exposed,      Limits,      Limits,
SIC    Production     Minimum      Maximum      Minimum      Maximum
Code      Workers    Estimate     Estimate     Estimate     Estimate
______________________________________________________________________

20       978,126      130,431      163,211       85,097       91,363
22       461,701      147,121      178,071       12,501       12,684
23       581,733       83,337      148,661       12,364       13,860
24       571,195      325,616      410,618      102,189      133,484
25       497,030      189,875      435,912       68,961       88,742
26       667,082      337,138      529,422       64,957       82,604
27     1,255,639    1,088,136    1,194,491      117,056      129,403
28       867,659      820,699      862,707      336,379      457,433
29       178,202      177,071      178,202      135,036      135,036
30       817,652      608,461      810,617      152,668      528,822
31       165,086       47,401       96,642        7,182        8,967
32       581,740      279,657      430,016       32,404       45,923
33       754,180      382,962      688,325      161,573      280,691
34     1,197,681      445,143    1,005,075      125,015      191,870
35     1,513,419      705,075    1,213,624      122,931      159,542
36     1,421,373      764,578    1,027,410      206,952      251,456
37     1,270,731      703,707      968,832       78,976      161,816
38       635,519      387,077      477,259       25,701       25,943
39       396,813      224,118      292,568       54,683       95,354
40       191,556      174,308      177,545        1,953        1,953
45       400,571      127,357      148,199            0            0
47        94,421       27,723       27,926          326          326
49       788,744      764,824      774,685      394,767      394,767
50     2,687,033      466,280      838,602      147,896      157,612
51       998,179      580,637      693,822      236,986      245,220
55     1,501,969    1,024,565    1,501,969      624,493      818,599
72       741,954      587,265      646,222       72,941       77,299
73     3,872,633    1,869,333    2,164,155      349,261      511,836
75       621,563      457,031      592,454       53,966       55,348
76       331,435      230,403      246,725       71,846      121,570
80     6,193,016    3,552,114    6,008,369       28,465       43,369
      __________    _________   __________    _________    _________
 Tot. 33,235,635   17,709,443   24,932,336    3,885,525    5,322,892
______________________________________________________________________

TABLE D-2.  Number of Workers Exposed and Number of Workers for
            Whom Risk is Reduced, by 2-Digit SIC - Continued
___________________________________________
        #Workers    # Workers    # Workers
            With         With         With
         Reduced      Reduced      Reduced
            Risk         Risk         Risk
SIC    (80% Risk    (90% Risk    (95% Risk
Code  Reduction)   Reduction)   Reduction)
___________________________________________
20        70,584       79,407       83,819
22        10,074       11,333       11,963
23        10,490       11,801       12,456
24        94,269      106,053      111,945
25        63,081       70,966       74,909
26        59,024       66,402       70,091
27        98,584      110,907      117,068
28       317,525      357,215      377,061
29       108,029      121,532      128,284
30       272,596      306,671      323,708
31         6,460        7,267        7,671
32        31,331       35,247       37,205
33       176,906      199,019      210,075
34       126,754      142,598      150,520
35       112,989      127,113      134,175
36       183,363      206,284      217,744
37        96,317      108,356      114,376
38        20,658       23,240       24,531
39        60,015       67,517       71,268
40         1,562        1,758        1,855
45             0            0            0
47           261          293          310
49       315,814      355,290      375,029
50       122,203      137,479      145,116
51       192,882      216,993      229,048
55       577,237      649,391      685,469
72        60,096       67,608       71,364
73       344,439      387,494      409,021
75        43,726       49,191       51,924
76        77,366       87,037       91,873
80        28,734       32,325       34,121
       _________    _________    _________
Tot.   3,683,369    4,143,787    4,373,999
___________________________________________

Employees exposed above the final limits are considered to be "at risk"

of adverse health effects. It should be noted that this presentation shows risk reduction in employee-equivalent terms; while all (100 percent) of the workers currently exposed above the new limits would benefit from reduced risk, the new lower limits would not eliminate all chemical exposure risk. An estimated five, ten, or twenty percent residual risk equivalent would remain at the new lower limits. Although not quantified, all employees currently exposed to hazardous substances at or below the recommended new levels would experience this residual risk. To obtain an approximation of risk reduction at the revised exposure levels, OSHA estimated that 95, 90, or 80 percent of the workers currently exposed above the limits (i.e., the midpoint between the minimum and maximum estimates) will benefit from reduced risk after their exposures are lowered to or below the final limits. The results of this analysis are also presented by two-digit SIC codes in Table D-2.

The American Iron and Steel Institute (AISI) [Ex. 72] objected to the approach used by OSHA to obtain the combined assessment. The AISI illustrated this point with the following example:

[T]he approach used by OSHA in obtaining [exposure estimates]...is not logical. If exposures to a substance are indicated in both the IMIS and survey results, the two numbers are averaged. If only one set reports exposures, then that set is used independently. Under this approach, the combined assessment for steelworkers exposed above the proposed standard for titanium dioxide is listed as 54,510 (the figure derived from OSHA's IMIS data alone) because no data was identified in this category in OSHA's telephone survey. It is not clear why the survey does not report data on titanium dioxide because according to OSHA the chemicals in the survey were "selected on the basis of...known exposure problems...." If the survey had determined there were zero workers overexposed, then the combined assessment would have been reduced to 27,255 - that is, 54,510 divided by 2 [Ex. 72, p. 11].

OSHA believes that it has designed an approach that makes optimum use of all of the exposure data available to the Agency. By using the averaging method described above, neither the survey data nor the IMIS data are given greater weight. OSHA realizes, as AISI points out, that the estimate of the number of employees exposed to a given chemical in a given industry sector is sensitive because OSHA's methodology uses an averaging approach. However, OSHA believes that the alternative, i.e., reliance on one data set as opposed to the other, is more disadvantageous because a vast amount of exposure information would have been ignored in the analysis. In addition, OSHA believes that making full use of both the IMIS and survey data minimizes any biases that may be inherent in the information contained in either data set alone. Furthermore, OSHA believes that, by determining maximum and minimum estimates of numbers of exposed employees, uncertainties in the analysis are appropriately recognized.

Some commenters [Exs. 3-890, 8-10, 8-31, and 8-32; Tr. 4-257] provided alternative estimates of the number of employees exposed and overexposed to specific chemicals in specific industries. For example, Dr. Boyd of the Styrene Information Research Council (SIRC) reports that, in the reinforced plastics industry segment, 30,000 workers are potentially exposed to styrene [Ex. 8-32, p. 1], a figure smaller than that estimated by OSHA. Eric Frumin, Director of Occupational Safety and Health for the Amalgamated Clothing and Textile Workers Union (ACTWU), reports that OSHA's estimates severely understate the number of workers potentially exposed to perchloroethylene in the dry cleaning industry [Ex. 8-31, pp. 21-23]. OSHA evaluated each of these commenter's estimates to determine whether OSHA's aggregate benefits estimates of exposed workers needed to be revised to reflect this new record evidence. Because OSHA received comparatively few data on the number of employees exposed, when compared with the number of industry sectors and substances covered in this analysis, OSHA has determined that incorporating the estimates provided by commenters would not substantially alter OSHA's aggregate estimate of the benefits associated with revising the air contaminant limits.

Estimates of the Reduction in Illness Cases and Lost Workdays

The BLS LABSTAT data base contains illness and lost-workday rates by SIC code. These rates are expressed as the annual number of illness cases or number of lost workdays per 100 full-time-equivalent employees. Reducing employee exposures to hazardous substances to a level below that associated with adverse health effects will result in a decrease in the number of illness cases and lost workdays.

To assess the impact on illness and lost-workday rates of reducing employee exposures, OSHA first examined the relationship between the percentage of workers estimated to be exposed above the final exposure limits for the 160 substances represented in the IMIS data base and current illness and lost-workday rates. This analysis was conducted at the three-digit SIC code level because of the lack of illness-rate data for some of the four-digit SIC code groups. The results of this analysis are presented graphically in Figure D-1. Among three-digit industries for which OSHA has found that no employees are currently exposed above the final limits, total illness case rates reported by the BLS for the same industry group are usually less than 0.2 cases per 100 employees per year, and frequently are reported to be zero. In contrast, where OSHA has found that an industry group has more than 10 percent of its workforce exposed above the final limits, total illness case rates above 0.2 case per 100 employees are frequently reported. In few instances does an industry group having 10 percent or more of its workforce exposed above the final limits report a total illness case rate of zero. Among three-digit SIC code industry groups for which OSHA has not found employee exposures above the final limits, 38 percent of the groups reported an illness rate of zero, 43 percent reported an illness rate of 0.1 to 0.2 case per 100 employees, and only 19 percent of the industry groups reported an illness rate greater than 0.2 case per 100 employees (but none above 0.5 case per 100 employees). Given this distribution of illness rates across these particular industry groups, it is concluded that industry groups in which employee exposures have been controlled to or below the final limits will have an illness rate approximating 0.1 case per 100 employees. It is believed that total illness cases at the three-digit level will be reduced to no more than 0.1 case per 100 employees after employee exposures are reduced to or below the final limits.


    Figure D-1.  Relationship Between Employee Exposures to Hazardous
                 Substances and Industry Illness Rates

(For Figure D-1, Click Here)

OSHA performed a similar analysis that also indicates that the rate of lost-workday illness cases will decline to a base rate of 0.05 cases per 100 employees and the annual rate of lost workdays will decline to 1 case per 100 employees after employee exposures are reduced to or below the final limits.

OSHA estimated the number of illness cases and lost workdays potentially avoided annually by applying current illness rates to the estimated number of production workers per three-digit SIC group; this yielded an estimate of the annual number of illness cases and lost workdays reported by each three-digit SIC code industry group. It was assumed that these industries would experience illness rates of 0.1 cases per 100 employees, 0.05 lost-workday illness cases per 100 employees, and 1 lost workday per 100 employees per year. Using this approach, OSHA estimated that promulgation of the final limits will potentially avoid 55,365 illness cases per year, 23,346 lost-workday illness cases per year, and 519,421 lost workdays caused by illnesses per year.

The movement to a 0.1 illness rate is presented as a best estimate supported by OSHA's interpretation of the relationship between chemical exposure levels and current industry illness rates. It may be argued, however, that if a 0.1 illness rate were achieved, the reduction in illnesses could not be credited exclusively to OSHA's rulemaking initiative, since some portion of the current BLS illness rate is made up of illnesses associated with exposures to hazardous agents or physical stress (e.g., radiation, noise, ergonomic stress).

While no claim is made that this rulemaking action will reduce illnesses related to these causes, OSHA believes that the benefit estimates related to this final rule of over 55,000 illnesses, over 23,300 lost-workday illnesses, and over 519,000 lost workdays avoided each year are reasonable. This is based on the finding that company records, upon which the BLS data are based, rarely show chronic illnesses caused by exposures to toxic substances [4, 6]. The potential level of underreported illnesses in the BLS series is illustrated in a recent report by Landrigan and Markowitz [8]. Using California physicians' reports of occupational illnesses, these authors estimated an occupational illness rate among New York State employees that was more than twice the BLS illness rate [8].

Mr. Frank T. Ryan, Vice President of the Rubber Manufacturers Association (RMA), addressed OSHA's use of the illness rate data, commenting that

there may be confounding factors, such as exposure to other substances, work practices, and especially non-occupational considerations, which affect the illness and lost workday rates....Because this fundamental issue is not addressed in the calculations, the RIA does not derive meaningful illness or lost workday data [Ex. 3-877, p. 22].

Similarly, Peter Hernandez, Vice President for Employee Relations of the American Iron and Steel Institute (AISI), commented that the PRIA provided no basis for the assumption that including non-chemical-related illnesses in the calculation was offset by the underreporting of illnesses in the BLS statistic [Ex. 72].

Landrigan and Markowitz [8] provided a breakout of illness causes in the 1984 BLS statistics. They reported that about 30 percent of all illnesses were caused by trauma or exposure to physical agents. As described above, they also reported that illness cases may be underreported by as much as a factor of 2. This was also reported in a recent article by Suruda and Emmett [9]. Therefore, the extent to which illness cases are underreported far outweighs the proportion of illness cases not attributed to chemical exposures. As such, OSHA believes that inclusion of non-chemically related illnesses in the assessment does not result in an overestimate of the annual number of illness cases avoided by the final rule.

Mr. Hernandez of the AISI also suggested that OSHA's estimates for the illness rates are overestimated because they include dermatitis cases. He noted that:

Because dermatitis is among the most common reported illnesses and is not associated with the employee airborne exposures, we believe OSHA's reliance on Bureau of Labor Statistics data overstates the benefits [Ex. 72, p. 13].

Although it is true that many of the reported cases of occupational illnesses are skin disorders, OSHA believes that reducing employee airborne exposures will contribute to a reduction in the number of cases of dermatitis. As a general rule, workplaces that have many cases of dermatitis are also more likely to use poor work practices and to be lacking in engineering controls; such facilities will have higher airborne exposures. On the other hand, a well-engineered facility with low airborne exposures generally also controls its employees' dermal exposures, and therefore has few, if any, cases of dermatitis. Therefore, OSHA believes that promulgation of these exposure limits for air contaminants will encourage the use of improved work practices, which will, in turn, reduce the incidence of dermatitis.

Estimates of the Number of Employees Potentially at Risk by Type of Hazard

In addition to estimating the number of employees exposed to the substances included in this analysis, OSHA also estimated the number of employees who are at risk of experiencing particular types of adverse health effects. To conduct this analysis, each substance included in the rulemaking was assigned to a health hazard category; these assignments were based on the primary health effects that provided the impetus for reducing a previous limit or establishing a new limit for a particular substance. (The assignment of substances to health effect categories is described in detail in Section VI-C of the preamble.) It should be noted that, in some instances, substances included in this rulemaking were grouped together in the preamble according to some basis other than a particular health effect; for example, several substances were grouped together because the ACGIH-recommended limits were derived based on the structural analogy of the grouped substances with that of other substances. For the benefits analysis described here, these substances were reclassified according to the primary health effect associated with exposure to the analogous chemical.

The number of employees estimated to be exposed to substances causing a particular health effect in an industry group was calculated by summing the number of employees exposed to all substances causing the same effect. Aggregate estimates across all affected industry sectors are presented in Table D-3. This table provides estimates of employees potentially exposed to substances in each health group, as well as estimates of employees exposed above the final limits for substances in each health group. Employees are frequently at risk from a variety of adverse health effects as a result of concurrent exposure to more than one toxic substance. Thus, the total number of employees considered to be at risk from any type of illness (as estimated in Table D-3) cannot be summed because the sum would result in doublecounting.


TABLE D-3.  Estimated Number of Workers Potentially at Risk of
            Experiencing Adverse Effects, by Type of Adverse Effect*
__________________________________________________________________________
                          # WORKERS     # WORKERS    #WORKERS     #WORKERS
                        POTENTIALLY   POTENTIALLY     EXPOSED      EXPOSED
                         EXPOSED TO    EXPOSED TO       ABOVE        ABOVE
                         SUBSTANCES    SUBSTANCES       FINAL        FINAL
                         ASSOCIATED    ASSOCIATED  LIMITS FOR   LIMITS FOR
                        WITH EFFECT   WITH EFFECT  SUBSTANCES   SUBSTANCES
ADVERSE HEALTH EFFECT     MIN. EST.     MAX. EST.   MIN. EST.    MAX. EST.
__________________________________________________________________________
PHYSICAL IRRITANT
  EFFECTS                 3,375,472     3,889,261     222,191      222,191
ODOR EFFECTS                519,318       521,938       3,597        3,597
SYSTEMIC TOXICITY         4,305,578     5,038,573     457,104      490,282
MUCOUS MEMBRANE
  IRRITATION             10,730,691    14,906,090     789,461    1,141,133
METABOLIC INTERFERENCES   4,015,702     4,205,530   1,233,413    1,241,564
LIVER/KIDNEY DISEASE      3,292,993     3,806,226     536,945      546,429
OCULAR DISTURBANCES       2,482,449     2,569,950      83,272      110,560
RESPIRATORY DISEASE       4,231,235     4,782,280   1,405,501    1,568,579
CARDIOVASCULAR DISEASE      166,077       166,868      44,403       44,403
NEUROPATHY                2,212,358     2,463,583     379,974      401,576
NARCOSIS                  6,966,024    10,520,982     941,472    1,073,717
CANCER                    1,712,799     1,851,342     465,013      528,650
ALLERGIC SENSITIZATION    2,545,551     2,648,973     305,955      305,955
__________________________________________________________________________
  Footnote(*)  Double counting of employees simultaneously exposed to more
than one substance in different adverse health effects categories prevents
the summation of workers exposed to all adverse health effects in this
table.

Estimates of the Number of Illness-Related Fatalities Avoided

As discussed in the preceding section, OSHA has estimated the number of employees currently at risk of experiencing a variety of adverse health effects brought about by overexposures to the substances included in this rulemaking. Many of these adverse effects, in particular, cancer, cardiovascular effects, chronic respiratory disease, and chronic liver and kidney damage, result in lethal outcomes. OSHA also believes that employees who are excessively exposed to substances causing systemic organ damage, neurological impairment, or metabolic effects (i.e., cardiovascular disease through excessive formation of methemoglobin or carboxyhemoglobin, and neurological impairment through cholinesterase inhibition) are at excess risk of incurring a fatal condition.

To estimate the number of fatalities associated with excessive exposure to the 212 substances included in this analysis, OSHA relied on standard U.S. mortality rates and on published estimates of the proportion of fatalities that are believed to be associated with occupational illnesses.

These data allowed OSHA to calculate cause-specific mortality rates that are attributable to occupational illnesses (i.e., mortality rates that represent the excess risk of mortality from occupational disease). OSHA then applied these occupationally related mortality rates to its estimates of the number of employees exposed to the substances of concern at levels above the final limits. OSHA's methodology and estimates are presented in Table D-4, and are described in detail below.


TABLE D-4. Estimated Annual Number of Fatalities Caused By Occupational
           Illness Among Workers Currently Exposed Above Final Limits,
           Using Alternative Assumptions
_________________________________________________________________________
        U.S. Annual    Total#      Number     Annual     Number  Annual #
         Death Rate    Deaths      Deaths  Dth. Rate    Workers    Fatal.
           /100,000     /Year      Attrib.  /100,000    Exposed     Among
Cause     Residents   in U.S.     to Occ.     Attrib.     Above      This
of        (1985) by        by     Ill. by   to Occup.     Final  Group of
Death      Cause(a)  Cause(b)       Cause    Illness.    Limits   Workers
_________________________________________________________________________
Cancer        193.3   461,484    23,074(c)     27.7     496,832(f)    138

Chronic        31.3    74,726     2,242(d)      2.7   1,487,040(g)     40
Pulmonary                           747(e)      0.9                    13
Disease

Chronic        11.2    26,739      802(d)       1.0      541,687(h)     5
Liver                              267(e)       0.3      541,687(h)     2
Disease

Cardio-
  vascular,   418.5   999,127    29,974(d)     35.9    2,146,360(i)   771
  Neurological,                   9,991(e)     12.0                   258
  and Renal

TOTAL, All Causes                                                 411-954
__________________________________________________________________________
  Footnote(a)  Source:  National Center for Health Statistics [5, Table
11].
  Footnote(b)  Based on a total residential population in 1985 of
238,740,000.[7, p. 18].
  Footnote(c)  Assumes 5 percent of all cancer deaths are of occupational
origin (Landrigan and Markowitz, 1987).
  Footnote(d)  Assumes 3 percent of all deaths are of occupational origin
Landrigan and Markowitz, 1987).
  Footnote(e)  Assumes 1 percent of all deaths are of occupational origin
(Landrigan and Markowitz, 1987).
  Footnote(f)  From Table D-3, midpoint estimate of number of workers
exposed above final limits for potential carcinogens.
  Footnote(g)  From Table D-3, midpoint estimate of number of workers
exposed above final limits for respiratory toxins.
  Footnote(h)  From Table D-3, midpoint estimate of number of workers
exposed above final limits for liver toxins.
  Footnote(i)  From Table D-3, midpoint estimate of number of workers
exposed above final limits for systemic toxins, metabolic toxins,
cardiovascular toxins, and neuropathic agents.

Estimate of the Number of Cancer Fatalities. The U.S. National Center for Health Statistics has published cause-specific U.S. mortality rates for 1985 (the most recent data available) [5]. This source reported that the annual U.S. cancer death rate in 1985 was 193.3 per 100,000 residents. Based on a total resident U.S. population of 238,740,000 in 1985 [7, p. 18], the number of cancer deaths that occurred in 1985 was 461,484. Landrigan and Markowitz [8] reviewed several published estimates of the percentage of cancer deaths that are attributable to occupationally related disease; these estimates range from less than 5 percent to 33 percent of all cancer deaths. Landrigan and Markowitz believe that, as a best estimate, 10 percent of all cancer deaths have an occupational origin.

Several commenters (Exs. 3-527 and 3-877) expressed the opinion that OSHA had overestimated the number of occupationally induced cancers that would be prevented by promulgation of the final rule's limits. Specifically, they criticized the study by Landrigan and Markowitz, who estimated that 5 to 33 percent of cancer deaths and 1 to 3 percent of all deaths have occupational origins. To support this argument, Mr. Ryan, of the RMA, cited the Doll-Peto study, which stated:

On present knowledge, it is impossible to make any precise estimate of the proportion of the cancers today that are attributable to hazards at work...and none of the estimates that have been made are claimed to be anything more than informed guesses....

...Until objective, nationally representative studies are undertaken, a more realistic assessment of the role of occupational hazards can probably be obtained by considering each type of cancer separately and estimating for each type the possible contribution of occupation.

...The proportion of cancer deaths that we have tentatively attributed to occupational causes is, therefore, about 17,000 out of 400,000; i.e., about 4 percent of all U.S. cancer deaths (Ex. 3-877, pp. 23-24).

At the informal hearing, Dr. Landrigan testified that his estimate that 10 percent of all cancers are occupationally induced is reasonable:

[T]he Doll-Peto estimate is low, for several reasons. First of all, they did not include in their estimate cancers which occurred in people over the age of 65.

Many occupational cancers don't develop in people until 20, 30, or even 40 years after exposure has occurred. Therefore, to cut off attribution of cancer to occupational exposure at age 65 almost certainly reduces the proportion of all cancers which can be attributed to occupation.

Another factor...which...diminishes the accuracy of the Doll-Peto estimate is that they excluded from consideration certain categories of cancer.

I think that 10 percent is a reasonable middle-of-the-road estimate. If you like arithmetic manipulation, then that figure is the geometric mean...between the Doll-Peto estimate of 4 percent and the old Califano estimate of 38 percent (Tr. p. 3-285).

OSHA believes that Dr. Landrigan's assessment of the Doll and Peto study is reasonable. Given the wide range in published estimates of the proportion of cancers that are occupationally related (4 to 33 percent), OSHA used alternative estimates of five and ten percent in the PRIA; the assessment for this final rule is based on the five percent estimate of all cancer deaths being occupationally related. Use of the five percent estimate is consistent with OSHA's recent benefits analysis for the Hazard Communication standard.

Using an occupational cancer death estimate of 5 percent and applying it to the estimated number of cancer deaths in 1985, OSHA estimates that 23,074 occupationally related cancer deaths occurred in the United States in that year (Table D-4).

As the next step, OSHA estimated the overall cancer death rate, both among the population that is occupationally exposed to chemicals and among the remainder of the population. In 1985, there were an average of 108,856,000 persons employed [10, p. 8]. However, 25,469,200 of these were employed in industries or occupations in which there is a low risk of exposure to toxic substances, such as finance, insurance, real estate, and private households [10, pp. 30, 84-88]. The remaining 83,386,800 persons are considered to be occupationally exposed to chemicals in varying degrees. Many would have only intermittent exposures at very low levels. Assuming that 5 percent of all cancer deaths are of occupational origin, OSHA calculated the annual cancer death rate attributed to occupational exposure by dividing the number of cancer deaths attributable to occupational illness by the population exposed, and multiplying that figure by 100,000. OSHA estimates that the annual cancer mortality rate attributable to occupational exposure to toxic substances is 27.7 per 100,000. OSHA then estimated that there are 496,832 workers currently exposed above the final limits to the potential carcinogens included in this rulemaking for which data were available. Applying the work-related cancer death rates to this population, OSHA estimates that 138 cancer fatalities occur each year among these workers, and that these fatalities will be prevented by the final rule.

In arriving at this estimate, two important offsetting arguments were considered. Because some of these workers may also be exposed to occupational carcinogens that are not covered in this rulemaking (such as asbestos or benzene), the number of occupational cancer deaths attributed to the substances included in this rulemaking may be overestimated. Offsetting this potential overestimate is the fact that the excess mortality rate of 27.7 per 100,000 workers was developed on the basis of occupational exposures among all workers. However, the excess mortality rate experienced among workers with high average exposures to hazardous chemicals typically runs at least two to three times higher than the national average rate. In consideration of this, OSHA believes that any overestimate of cancer fatalities avoided attributed to regulated chemicals not covered under this rulemaking is offset by the use of a mortality rate that understates the true excess mortality rate among workers with very high exposures to toxic chemicals. (Additional comments on excess mortality rate estimates are included in the final section of this chapter.) An alternative analysis of the reduction in cancer mortality was conducted using OSHA's quantitative risk assessments for the potential human carcinogens included in this rulemaking (the results of OSHA's risk assessments are presented in the preamble to the final rule). This analysis is presented in Table D-5. Using available data from the combined IMIS and 1988 survey, OSHA found that employees are currently exposed above the final limits to four of the 17 potential carcinogens listed (acrylamide, carbon tetrachloride, chloroform, and perchloroethylene). Applying quantitative risk estimates to the estimated number of workers currently overexposed to these four substances only, OSHA estimates that compliance with the final limits will avoid 11,519 cancer fatalities over the working lifetime of the population (i.e., 45 years). The average annual reduction in the number of cancer fatalities avoided over 45 years is estimated to be 256.


TABLE D-5.  Estimates of Cancer Deaths Potentially Avoided, Based
            on Quantitative Risk Assessments,(a) Over 45 Years
________________________________________________________________
                       Number of                       Estimated
                         Workers       Estimated       Number of
                           Above       Number of          Cancer
                           Final          Cancer          Deaths
Substance                  Limit          Deaths         Avoided
________________________________________________________________

Acrylamide                 7,896              79              71
Carbon
  Tetrachloride           97,134           1,739           1,380
Chloroform               123,950           2,776           2,743
Perchloroethylene        267,821          12,052           7,325

TOTAL, 45 years                           16,646          11,519
________________________________________________________________
ANNUAL                                       370             256
________________________________________________________________
  Footnote(a) Risk assessments are presented in Section IV-C-15 of the
preamble. The assessment for perchloroethylene is based on risk estimates
developed by Dr. Dale Hattis (Ex. 8-31, App. 11-A).

As noted, although OSHA has evaluated the cancer risk for 17 potential carcinogens, there were IMIS data, survey data, and quantitative risk assessments (all of which are necessary for benefits analysis) for only four of these. Lack of IMIS or survey data means that the substance has not been sampled by an OSHA compliance officer or that none of the survey participants indicated that the substance was used at their facilities. This does not mean that no workers are currently exposed to these substances. Lacking a basis for estimating the extent of employee exposure, OSHA could not estimate the extent of reduction in cancer deaths attributable to the reduction in exposure limits for these substances. To the extent that employee exposure to these carcinogens is reduced, further reductions in the number of cancer deaths will occur.

Estimated Reduction in Occupational Deaths from Causes Other than Cancer. As shown in Table D-4, OSHA also estimated the number of occupationally related fatalities that are expected to occur annually among employees exposed to substances associated with adverse health effects other than cancer. To perform this analysis, OSHA relied on an estimate made by Landrigan and Markowitz [7] that between 1 and 3 percent of all nonmalignant disease is of occupational origin. Using the 1- and 3-percent figures as alternative assumptions and using the same methodology as that described above for cancer deaths, Table D-4 shows the following:

- Between 13 and 40 deaths caused by respiratory disease are estimated to occur each year among workers exposed to respiratory toxins covered in this rulemaking;

- Between 2 and 5 deaths are estimated to occur each year among workers exposed to liver toxins covered in this rulemaking; and

- Between 258 and 771 deaths are estimated to occur each year among workers exposed to systemic toxins, cardiovascular toxins, metabolic toxins, and neurological toxins covered in this rulemaking.

Summing these estimates, OSHA believes that between 411 and 954 non-cancer-related occupational fatalities occur each year. The same offsetting considerations discussed in the analysis of the cancer fatalities avoided under this rule also apply here. While some substances are being controlled by activity outside of this rulemaking, any overestimation effect is balanced by an underestimate of the real excess mortality rate for workers with high exposure levels to the chemicals under consideration.

The Chemical Manufacturers Association (CMA) (Ex. 3-527) stated that OSHA made the assumption that reducing exposures will eliminate all cancer fatalities that are estimated to occur from exposure to carcinogens included in the rulemaking. They argue that this is inconsistent with OSHA's statement in the preamble that

there is reason to assume that the dose-response of most carcinogens will follow a linear, non-threshold relationship (Preamble 21190. Col. 3) [Ex. 3-527, p. 42].

OSHA's approach, based on estimating excess death rates (Table D-4), did assume that all cancers caused by exposure to the four substances would be avoided; however, changing this assumption would not have a major impact on the estimated total number of fatalities avoided. For example, if it is assumed that only half of the estimated number of cancer fatalities would be avoided (a conservative assumption, given that most PELs are being reduced by more than a factor of 5), then the estimated annual number of cancer fatalities avoided would be 69 rather than 138. The estimated total annual number of avoidable deaths from all causes would range from 342 to 885. This estimate is only about 7 to 17 percent less than the estimate of 411 to 954 avoidable deaths reported in the Table D-4. Since cancer fatalities avoided represent only a part of the benefits to be achieved through this rulemaking, changing the assumption on cancers avoided will not result in a substantial change to the total number of avoidable fatalities attributable to revising the PELs. Furthermore, OSHA's alternative approach, which relies on the quantitative estimates of risk, identified a larger number of cancers avoided each year (256). This latter method takes into account the presence of residual risk at the revised PELs.

AISI (Ex. 188, p. 43) also argued that OSHA overstated benefits estimates presented in the PRIA because the effect of the Hazard Communication standard was not considered. In that rulemaking, OSHA determined that the hazard communication standard could reduce occupation-related cancers by 20 percent. If the beneficial effects of the Hazard Communication standard are considered in assessing the benefits associated with revising the PELs on Table Z, the maximum effect would be to reduce OSHA's estimate of cancer fatalities avoided by 20 percent (i.e., from 138 to 115).

In sum, the combined estimate for the number of cancer and noncancer deaths potentially avoided each year by compliance with the new limits is between 411 and 954 or an average of 683 fatalities avoided each year. OSHA considers these to be reasonable estimates of the benefits associated with revising the PELs on Table Z.

Additional Comments and an Alternative Method for Estimating Excess Mortality Rates. The analysis described above to estimate the number of fatalities that are potentially preventable relies on published estimates of the proportion of all U.S. fatalities that are believed to result from occupational illnesses. These estimates were used with U.S. cause-specific mortality rate figures to estimate the excess mortality rate among all U.S. workers, by cause of death (shown in Table D-4).

In making these excess mortality rate estimates, OSHA applied the excess number of fatalities across the entire U.S. working population. Implicit in this approach is an assumption that all workers are at some risk of fatality from all causes of death. In fact, only a portion of the workforce is at risk of fatality from each type of occupational illness. Deaths will occur only among workers who are potentially exposed to carcinogens; no excess deaths will occur among workers who are not so exposed. Similarly, not all workers are at risk of dying from occupationally related cardiovascular illnesses; only some portion of the workforce are at excess risk, and all fatalities resulting from occupationally related cardiovascular disease will occur among this subset of workers. Because OSHA's excess mortality rate estimates presented earlier were derived by applying the estimated number of work-related fatalities across the entire U.S. workforce, excess mortality rate figures are likely to be substantially understated.

To assess the magnitude of this bias, OSHA conducted an alternative analysis to estimate the number of work-related fatalities that are expected to occur among workers exposed above the final limits. This alternative assessment relied on judgments regarding the general increase in mortality rates that are frequently observed in epidemiologic studies that demonstrate a causal relationship between exposure to toxic substances and excess disease mortality. The alternative assessment is presented in Table D-6.


TABLE D-6. Alternative Assessment of Number of Fatalities Expected to
           Occur Among Workers Currently Exposed Above Final Limits

_________________________________________________________________________
                  United States   Est. Excess   Number of
                 Cause-Specific    Mort. Rate     Workers   Annual Number
                 Mortality Rate      /100,000     Exposed   of Fatalities
                       /100,000    Workers at       Above      Among This
                   Residents(a)     Risk From       Final        Group of
Cause of Death        (1985)           Hazard      Limits         Workers
_________________________________________________________________________

Cancer                    193.3      193.3(b)    496,832(c)           960

Chronic Pulmonary          31.3        9.4(d)  1,487,040(c)           140
Disease

Chronic Liver              11.2        3.3(d)    541,687(c)            18
Disease

Cardiovascular,           418.5      125.6(d)  2,146,360(c)         2,696
                                                                    _____
TOTAL                                                               3,814
_________________________________________________________________________
  Footnote(a)  Source: National Center for Health Statistics [5, Table 11].
  Footnote(b)  Assumes that overall cancer mortality rate among workers at
risk is twice the U.S. rate (i.e., a 100-percent excess rate).
  Footnote(c)  From Table D-4.
  Footnote(d)  Assumes that overall disease mortality rate among workers
at risk is 1.3 times the U.S. rate (i.e., a 30-percent excess risk).

The overall U.S. cancer mortality rate for 1985 is 193.3 deaths per 100,000 residents (Table D-6). Typically, when causal relationships between exposure and excess lung cancer mortality are found in epidemiologic investigations, the exposed cohort frequently shows a cancer mortality rate of 1.1 to 10 times higher than the general population. For cancers that are more rare than lung cancer, mortality rates among working populations may be 50 times higher than for the general population. An alternative estimate of the number of cancer fatalities expected to occur among the estimated 496,832 workers exposed above the final limits for the potential carcinogens could be developed based on the assumption that the overall cancer fatality rate among these workers is twice that of the U.S. population (i.e., 386.6 per 100,000 workers versus 193.3 per 100,000 residents). The excess cancer mortality rate among these workers is therefore assumed to be 193.3 per 100,000 workers (386.6 minus 193.3). Applying this estimated excess cancer mortality rate to the 496,832 workers exposed above the final limits yields an estimated 960 cancer deaths occurring annually that are attributable to occupational exposure. For example, Duh and Asal [Ex. 8-31, App. 7] reported excess lung cancer death rates for drycleaning workers of 1.7 times expected death rates, as well as kidney cancer death rates of 3.8 times expected rates. Similarly, EPA [Ex. 1-1132] reported standard mortality ratios for lung cancer of generally between one and ten for nickel refinery workers. This same approach could be used for estimating non-cancer-related fatalities assuming that the overall fatality rate among workers at risk from these illnesses is 1.3 times the corresponding U.S. mortality rate (mortality rates of 1.1 to 1.5 are frequently observed in epidemiologic studies demonstrating causal relationships between exposure and excess fatalities). This amounts to an excess mortality rate of 30 percent above the overall U.S. rate. Applying these excess mortality rate figures to the estimated worker populations exposed above the final limits, OSHA estimates that, among these workers, 140 deaths occur annually due to chronic pulmonary disease, 18 deaths occur annually due to liver disease, and 2,696 deaths occur annually due to cardiovascular, neurological, and renal diseases. In total, including cancer, OSHA estimates that 3,814 work-related fatalities (including those from cancer) may be occurring each year among employees who are exposed above the limits to the hazardous substances included in this rulemaking.

References

1. U.S. Department of Labor. Bureau of Labor Statistics. LABSTAT data base (unpublished data). 1985.

2. Dun and Bradstreet, Inc. Dun's Market Identifiers (database). 1985.

3. U.S. Department of Commerce. Bureau of the Census. County Business Patterns (database). 1985.

4. Office of Technology Assessment. Preventing Illness and Injury in the Workplace. 1985.

5. U.S. Department of Health and Human Services. National Center for Health Statistics. Advance Report of Final Mortality Statistics, 1985. Monthly Vital Statistics Report: 36, No. 5, Supplement. August 28, 1987.

6. U.S. Department of Labor. An Interim Report to Congress on Occupational Diseases. 1980.

7. U.S. Department of Commerce. Bureau of the Census. Statistical Abstract of the United States. 1987.

8. Landrigan, P.J. and S.B. Markowitz. Occupational Disease in New York State - Report to the New York State Legislature. February 1987.

9. Suruda, A. and Emmett, E.A. Counting recognized occupational deaths in the United States. 1988. Journal of Occupational Medicine 30:868-872.

10. U.S. Department of Labor. Bureau of Labor Statistics. Employment and Earnings. July 1986.

E. Assessment of Nonregulatory Alternatives

Introduction

The declared purpose of the Occupational Safety and Health (OSH) Act of 1970 is "...to assure so far as possible every working man and woman in the Nation safe and healthful working conditions and to preserve our human resources...." Thus, the Act requires the Secretary of Labor, when promulgating occupational safety and health standards for toxic materials or harmful physical agents, to set the standard "...that most adequately assures, to the extent feasible, on the basis of the best available evidence, that no employee will suffer material impairment of health or functional capacity...." It is on the basis of this congressional directive that OSHA has initiated regulatory actions to reduce the adverse health effects associated with occupational exposure to hazardous substances.

Market Failure

Economic theory suggests that the need for government regulation is greatly reduced where private markets work efficiently and effectively to allocate health and safety resources. The theory typically assumes perfectly competitive labor markets where workers, having perfect knowledge of job risks and being perfectly mobile among jobs, command wage premiums that fully compensate for any risk of future harm. Thus, theoretically, the costs of occupational injury and illness are borne initially by the firms responsible for the hazardous workplace conditions and, ultimately, by the consumers who pay higher prices for the final goods and services produced by these firms. With all costs internalized, private employers have an incentive to reduce hazards wherever the cost of hazard abatement is less than the cost of the expected injury or illness. The resultant level of safety and health is considered "efficient" in the sense that it minimizes the sum of the costs of hazard prevention and of injury or illness. Perfectly competitive labor markets, however, do not exist for many industrial markets. OSHA, therefore, believes that it must take appropriate actions to provide greater health protection for workers exposed to toxic substances.

Evidence indicates that market forces have not been effective in reducing excessive occupational exposure to hazardous substances, thereby contributing to the consequent development of occupational diseases. In spite of the danger associated with the inhalation or other exposure to hazardous substances, the social costs of production have not been internalized, in part, because of market imperfections and the existence of externalities. Consequently, the amount of protection that the private market will offer to workers differs from the socially desired level.

First, evidence on occupational health hazards in general suggests that in the absence of immediate or clear-cut danger, employees and employers have little incentive to seek or provide information on the potential long-term effects of exposure. Employers faced with potentially high compensatory payments may, in fact, have a disincentive to provide information to employees. When relevant information is provided, however, employers and employees might still find informed decisionmaking a difficult task, especially where long latency periods precede the development of chronic disabling disease. Moreover, if signs and symptoms are nonspecific--that is, if an illness could be job-related or could have other causes--employees and employers may not link disease with such occupational exposure.

Second, even if workers were fully informed of the health risks associated with exposure to hazardous substances, many face limited employment options. Nontransferability of occupational skills and high national unemployment rates sharply reduce a worker's expectation of obtaining alternative employment quickly or easily. A worker employed in a foundry, for example, could find it difficult to apply occupational skills to a new job in searching for a safer workplace.

In many regions of the country, the practical choice for workers is not between a safe job and a better paying but more hazardous position, but simply between employment and unemployment at the prevailing rates of pay and risk. In addition to the fear of substantial income loss from prolonged periods of unemployment, the high costs of relocation, the reluctance to break family and community ties, and the growth of institutional factors such as pension plans and seniority rights serve to elevate the cost of job transfer. Thus, especially where wages are more responsive to the demands of more mobile workers who tend to be younger and perhaps less aware of job risks, hazard premiums for the average worker will not fully compensate. Where this is the case, labor market negotiations are unlikely to reflect accurately the value that workers place on health.

In addition to the market imperfections, externalities occur if employers and employees settle for an inefficiently low level of protection from hazardous substances. For the competitive market to function efficiently, only workers and their employers should be affected by the level of safety and health provided in market transactions. In the case of occupational safety and health, however, society shares part of the financial burden of occupationally induced diseases, including the costs of premature death, chronic illness, and disability. Those individuals who suffer from occupationally related illness are cared for and compensated by society through taxpayer support of social programs, including welfare, Social Security, and Medicare. These combined factors of labor market imperfections and the existence of externalities contribute to the failure of the market to supply healthful working conditions in industries where hazardous substances exist.

Tort Liability

The use of liability under tort law is one nonregulatory alternative that has been increasingly used in litigation concerning occupationally related illnesses. Prosser [1] describes a tort, in part, as a "civil wrong, other than a breach of contract, for which the court will provide a remedy in the form of an action for damages," although he says that "a really satisfactory definition has yet to be found."

If the tort system applies, it would allow a worker whose health has been adversely affected by occupational exposure to a hazardous substance to sue and recover damages from the employer. Thus, if the tort system is effectively applied, it might shift the liability of direct costs of occupational disease from the worker to the firm under certain specific circumstances.

With very limited exceptions, however, the tort system is not a viable alternative in dealings between employees and their employers. All states have legislation providing that Workers' Compensation is either the exclusive or principal remedy available to employees against their employers. Thus, under tort law, workers with an occupational disease caused by exposure to a hazardous substance can only file a product liability suit against a third party manufacturer (e.g., Johns Manville), processor, distributor, sales firm, installer, agency, or contractor. It is often difficult, however, to demonstrate a direct link between an exposure to a hazardous substance and the illness.

In order to pursue litigation successfully, there must be specific knowledge of the magnitude and duration of a worker's exposure to a hazardous substance, as well as the causal link between the disease and the occupational exposure. Usually, it is extremely difficult to isolate the role of occupational exposures in causing the disease, especially if workers are exposed to many toxic substances. This difficulty is further compounded by the long latency periods that are frequently involved. In addition, the liable party must be identifiable, but workers may have several employers over a working lifetime. The burden of proof that an occupational exposure to a hazardous substance occurred, that a specific employer is the liable party, and that the exposure level was significant may prohibit the individual from initiating the suit.

The costs associated with producing information and with litigation itself may be quite substantial. First, information is a public good, which means that once produced it can be transmitted inexpensively to any number of individuals without diminishing the quality or quantity of the information. It is, therefore, difficult to control distribution once the information is produced. A producer of information may find that information produced at great expense can be acquired freely by potential customers, and that consequently, the market for the information has virtually disappeared. As a result, public goods are typically underproduced relative to what is considered economically efficient. This general undersupply of information adversely affects workers' awareness of the cause of their illness and thus reduces the likelihood that they will pursue tort liability suits.

Second, legal proceedings impose costs on both plaintiffs and defendants. In deciding whether to sue, the tort victim must be sure that the size of the claim will be large enough to cover legal expenses. In effect, the plaintiff is likely to face substantial transaction costs in the form of a contingency fee, commonly 33 percent, plus additional legal expenses. The accused firm must also pay for its defense.

The majority of occupational disease tort activity has involved workers exposed to asbestos. To date, approximately 100,000 individual plaintiffs have filed asbestos lawsuits in the country. These employees avoided the exclusive remedy of Workers' Compensation by suing suppliers of asbestos instead of employers. A report prepared by the Research Triangle Institute entitled, Tort Liability and Worker Health: An Examination of the Economic, Legal, and Scientific Issues Surrounding the Occupational Disease Protection Afforded by Tort Law [2], contains some data pertaining to legal costs and the size of awards. One investigator, for example, found that an average ratio of legal costs to proceeds was 37 percent for a sample of cases. The data, however, do not separate legal fees paid by the defendants and plaintiffs.

Insurance and liability costs are not borne in full by the specific employer responsible for the risk involved. For firms that are insured, the premium determination process is such that premiums only partially reflect changes in risk associated with changes in exposure to hazardous substances. This lack of complete adjustment is the so-called "moral hazard problem," which is the risk that arises from the possible dishonesty or imprudence of the insured. As the insured firm has paid an insurance company to assume some of the risks, that firm has less reason to exercise the diligence necessary to avoid losses. Transfer of risk is a fundamental source of imperfection in markets(1).


__________
  Footnote(1) For a general discussion of moral hazard as a source of
market failure, see Arrow [4] and Spence and Zeckhauser [5].  For
applications of this concept to employee health and safety, see Chelius
[6], Rea [7], and Consad and General Research Corporation [8, Section
5.1].

For firms that self-insure or carry liability insurance with a large deductible, the costs of a single claim may be fully borne by the firm. Very small firms, and large firms with a large number of claims, however, may fail to meet the full costs by declaring bankruptcy. For example, the Johns Manville Corporation(2) declared bankruptcy to avoid massive claims associated with asbestos-related disease. Although the firm experienced a sharp decline in the value of its stock, it is still in business, while its obligation to pay asbestos-related claims is in considerable doubt. Other asbestos producers, including U.N.R. Industries, Inc. and Amatex Corporation, have followed the example of the Manville Corporation by filing for bankruptcy [9], further reducing the chances that their workers or others who contract asbestos-related diseases will collect Workers' Compensation or tort liability awards.


__________
  Footnote(2) Johns Manville Corporation, formerly the world's largest
asbestos manufacturer, filed for Chapter XI protection under the Federal
Bankruptcy Law in August 1982.  The company was financially solvent when
it filed for bankruptcy but estimated that it would ultimately face a cost
of more than $2 billion to settle 52,000 asbestos-related claims.  In the
meantime, the company's assets have been frozen and successful plaintiffs
cannot collect awards [9].

Workers' Compensation

The Workers' Compensation system is a result of the perceived inadequacies in liability or insurance systems to compel employers to prevent occupational disease or compensate workers fully for their losses. The system was designed to internalize some of the social costs of production, but in reality, it has fallen short of compensating workers adequately for occupationally related disease. Thus, society shares the burden of occupationally related adverse health effects, premature mortality, excess morbidity, and disability through taxpayer support of social programs such as welfare, Social Security disability payments, and Medicare.

Compensation tends to be inadequate, especially in permanent disability cases, in view of the expiration of benefit entitlements and the failure to adjust benefits for changes in a worker's expected earnings over time. As of January 1987, 8 states still restricted permanent disability benefits either by specifying a maximum number of weeks for which benefits could be paid or by imposing a ceiling on dollar payments [10].

At present, time and dollar restrictions on benefit payments are even more prevalent in the area of survivor benefits. The duration of survivor benefits is often restricted to 10 years, and dollar maximums on survivor payments range from $7,000 to $60,000. In addition, it should be noted that if the employee dies quickly from the occupational illness and has no dependents, the employer need pay only nominal damages under Workers' Compensation (i.e., a $1,000 death benefit).

Finally, in spite of current statutory protection, disability from occupational diseases represents a continuing, complex problem for Workers' Compensation programs. Occupational diseases may take years to develop, and more than one causal agent may be involved in their onset. Consequently, disabilities resulting from occupationally induced illness often are less clearly defined than those from occupationally induced injury. As a result, Workers' Compensation is often a weak remedy in the case of occupational disease. For example, as recently as April 1983, the U.S. Supreme Court refused to hear an occupational disease case (Richard D. Bunker v. National Gypsum Co.) involving a worker who was diagnosed as having asbestosis 23 years after the expiration of the 3-year time limit allowed by Indiana law for filing a compensation claim [11]. Indeed, there is some evidence indicating that the great majority of occupationally induced illnesses are never reported or compensated [12].

The insurance premiums paid by a firm under the Workers' Compensation system are generally not experience rated - that is, they do not reflect the individual firm's job safety and health record. About 80 percent of all firms are ineligible for experience rating because of their small size. Such firms are class rated, and rate reductions are granted only if the experience of the entire class improves. Even when firms have an experience rating, the premiums paid may not accurately reflect the true economic losses. Segregation of loss experience into classes is somewhat arbitrary, and an individual firm may be classified with other firms that have substantially different normal accident rates. An experience rating is generally based on the benefits paid to workers, not on the firm's safety record. Thus, employers may have a greater incentive to reduce premiums by contesting claims than by initiating safety measures.

In summary, the Workers' Compensation system suffers from several defects that seriously reduce its effectiveness in providing incentives for firms to create safe and healthful workplaces. The scheduled benefits are significantly less than the actual losses to the injured workers, and recovery is often very difficult in the case of occupational diseases. Thus, the existence of a Workers' Compensation system limits an employer's liability significantly below the actual costs of the injury. In addition, premiums for individual firms are unlikely to be specifically related to that firm's risk environment. The firm, therefore, does not receive the proper "signals" and consequently fails to invest sufficient resources in reducing workplace injuries and illnesses. The economic costs not borne by the employer are borne by the employee or, as is often the case, by society through public insurance and welfare programs.

Standards of Other Organizations

Traditionally, representatives of professional organizations have collectively developed voluntary guidelines to assist members in maintaining safe and healthful working conditions for their employees. These guidelines are widely disseminated among members of the organizations and, at times, have been adopted as guidelines by organizations beyond the initiating one as well as by industry groups. In some cases they have become the de facto industry standard. Three professional organizations have developed voluntary guidelines in the form of exposure limits for chemical substances: The American National Standards Institute (ANSI); the American Industrial Hygiene Association (AIHA); and the American Conference of Governmental Industrial Hygienists (ACGIH). ANSI has withdrawn its earlier hazardous substance standards and has stated it does not intend to publish any others. The AIHA has a rather limited list of recommended limits. However, the ACGIH has published an extensive list of threshold limit values (TLVs) for many years. The ACGIH is recognized throughout the world for its members' expertise and contribution to industrial hygiene.

In May 1971, OSHA adopted as Federal health standards the exposure limits recommended by ANSI and ACGIH for 425 chemicals. Since that time, advances in scientific knowledge have demonstrated that those limits are not always adequate to protect employee health. Consequently, the ACGIH, the professional organization which continues to develop TLVs, has changed its recommendations yearly to reflect later information. However, adherence to the TLVs developed after 1971 is purely voluntary. Except for imminent hazards, there is no sanction for failure to comply with the limits and many employers have not adopted practices which would control employee exposure to these new levels.

In addition to professional organizations, international bodies such as the European Economic Community, the International Labor Organization, and the World Health Organization have recommended exposure limits for some hazardous substances. While these limits may not be as widely known in the United States as those of U.S. professional organizations, they are made available to the industrial hygiene community through professional journals and meetings. Within the U.S., the National Institute for Occupational Safety and Health (NIOSH) of the Department of Health and Human Services has published recommended exposure limits (RELs) for a number of chemicals. These are publicized through NIOSH Current Intelligence Bulletins and other publications which are widely disseminated.

Although the ACGIH TLVs and the NIOSH RELs are widely recognized by health professionals and employers alike, OSHA has found that some employers are not complying voluntarily with the newer TLVs, the RELs, or the standards of other bodies. Chapter D discussed OSHA's estimates of the extent of exposures in excess of the TLVs, and the adverse health effects resulting from such exposure. OSHA believes that significant numbers of employees are exposed to chemicals at levels exceeding those recommended by other organizations, and that OSHA cannot rely on employers to comply voluntarily with the recommendations. Therefore, OSHA concluded that this nonregulatory alternative is not generating the optimal level of occupational health.

Conclusion

OSHA believes that there are no nonregulatory alternatives that adequately protect workers from the adverse health effects associated with exposure to the chemicals regulated in this rulemaking. OSHA believes that tort liability laws and Workers' Compensation do not provide adequate worker protection due to market imperfections. Some employers have not complied with the standards recommended by professional organizations. The deleterious health effects resulting from continued high levels of exposure to hazardous substances require a regulatory solution.

The National Grain and Feed Association (NGFA) has disagreed with OSHA's conclusion that this rule is necessitated by a situation of market failure [Ex. 180A]. NGFA claims that OSHA "ultimately rejects each of the alternatives because of what it characterizes as imperfections in the ability of each of the alternatives to meet fully that alternative's theoretical objectives." OSHA is not implying that non-regulatory alternatives are complete failures, but that they are not total successes. That they are partial failures is precisely the situation that creates the need for OSHA. Because OSHA cannot write and enforce a unique set of regulations for each facility, the regulations must be written and enforced on an industry-wide basis. This does not imply that all firms have failed to adopt guidelines, but that some have and that workers at these firms are potentially at risk.

The rules will not impact on those entities which have already adopted the voluntary guidelines of ACGIH and NIOSH. What the rule will do is compel those firms that have done little voluntarily, to act. Furthermore, firms which comply voluntarily can be at a competitive disadvantage in the short run. When some firms don't comply with voluntary standards, the pressure not to comply increases on all firms. When all firms must comply with a regulation, none should be at a competitive advantage or disadvantage as a direct result of the regulation.

NGFA also states that although none of the nonregulatory alternatives are perfect, imperfection is not justification for "dismissing that alternative as a failure," and that "the relevant question is whether or not these alternatives - on the whole - promote workplace safety, and to what extent they do so." OSHA fully agrees with these statements. Certainly the alternatives, when followed, do promote workplace safety, but the extent to which they do so may not be sufficient. The 1988 sample survey showed that, among firms where there are chemical exposures, 26 percent would not be in compliance with the new standards and less than 15 percent of firms making or using hazardous substances, did exposure monitoring. OSHA believes these facts reflect the market's failure to better control exposures to hazardous substances.

The situation of imperfect information is also questioned by NGFA. They cite the availability of information from various sources (news media, labor unions, local public interest organizations, plaintiffs' attorneys), stating "all of whom - for their own reasons - aggressively spread the word about substances that may present occupational health risks." The comments and testimony received for this rulemaking present an ideal example of the problems of information. First, the information is often presented in a selective manner precisely because the presenters are working "for their own reasons." The NGFA maintained in the hearing that there are no substantive hazardous exposure problems at grain elevators and has sued OSHA to try to prevent the hazard communication standard from informing their workers of any risk. Yet grain dust has been known to cause disease since at least 1713 and most, if not all, impartial expert organizations have concluded it does cause disease. Second, the sheer volume of information in some cases is overwhelming. Third, the highly technical nature of much information makes analysis extremely difficult, except for the specialist. For example, the industrial structure of grain handling is segmented among several major industry categories (SICs). Information from one subcategory can be inadvertently misrepresented to apply to all grain handling facilities. The combination of these factors can make analysis of information, even when it is available, very difficult or inconclusive.

NGFA asserts that the tort system provides better recourse for employees than OSHA admits. As just discussed, the tort system provides only an imperfect remedy. The employees' damages are restricted under Workers' Compensation and it is difficult to prove causation. The only possible defendant is the supplier, not the employer. This does not encourage employers to take precautions. The greatest problem with the tort system is that torts are a retroactive remedy, after illness or death, whereas OSHA has a responsibility to assure, to the extent feasible, that no employee will suffer material impairment of health or functional capacity prospectively. Although the threat of a tort may help to prevent health damage to employees, it remains more a form of compensation for injuries suffered than a preventive measure.

NGFA contends that, "A profit-maximizing employer certainly will incorporate those additional costs (insurance, hiring, training, goodwill) in its consideration of necessary safety measures." OSHA agrees that, in the ideal, employers and manufacturers would provide a high level of safety and health protection to their employees. This is not, however, reflected everywhere in reality. Long-term implications of safety and health problems are often ignored or underestimated in the pursuit of short-term profits.

OSHA, therefore, does not agree with the NGFA's arguments and continues to believe that there are no nonregulatory alternatives that adequately protect workers from the adverse health effects associated with exposure to the chemicals regulated in this final standard.

References

1. Prosser, William Lloyd. Handbook of the Law of Torts. 4th ed. St. Paul: West Publishing Company, 1971. 1208 Pp.

2. Morris, G.E. Tort Liability and Worker Health: An Examination of the Economic, Legal and Scientific Issues Surrounding the Occupational Disease Protection Afforded by Tort Law, Final Report. Prepared for the U.S. Department of Labor, Occupational Safety and Health Administration, Office of Regulatory Analysis. Research Triangle Park, North Carolina: Research Triangle Institute, 1982.

3. Crandall F. vs. Eureka Fluid Works U.S. District Court, D. Arizona, No. 61V 82-061-TUC-RMB. September 18, 1984.

4. Arrow, K.J. Essays in the Theory of Risk-Bearing. Chicago:

Markham Publishing Company, 1971. 278 Pp.

5. Spence, M., and Zeckhauser, R. "Insurance, information, and individual action" American Economic Revue. 61:380-387, 1971.

6. Chelius, J.R. Workplace Safety and Health: The Role of Workers' Compensation. Washington, DC: American Enterprise Institute for Public Policy Research, 1977.

7. Rea, S.A. Jr. "Workmen's compensation and occupational safety under imperfect information", American Economic Revue. 71:80-93, March 1981.

8. Consad and General Research Corp. Employer Compensation and Control Systems, Final Report, 1982. Prepared for the U.S. Department of Labor, Occupational Safety and Health Administration. Pittsburgh: Consad Research Corporation; and McLean, Virginia: General Research Corporation, 1982.

9. "Manville's bankruptcy ploy" Dun's Business Month, October 1982. 10. 1987 Analysis of Workers Compensation Laws. Prepared by the U.S.

Chamber of Commerce. Washington, D.C., 1987.

11. Young, L.R. "Job-related disease case refused" J Commerce. April 19, 1983.

12. Discher, David P.; Kleinmann, G.G.; and Foster, F.J. National Occupational Hazard Survey-Pilot Study for Development of an Occupational Disease Surveillance Method. Report No. NIOSH-75-162. Sponsored by the National Institute for Occupational Safety and Health, Department of Environmental Health. Seattle: University of Washington, May 1975.

F. Technological Feasibility

Feasibility Determination

This chapter presents a technological feasibility analysis of industry's ability to meet OSHA's final rule permissible exposure limits (PELs) for a wide range of occupational health hazards. These PELs would include limits on airborne concentrations of substances, and in some instances, direct contact of the skin with the substance.

The control of workplace exposures to toxic chemicals involves combining a variety of standard techniques to solve a situation-specific problem. OSHA believes that existing engineering controls are available to reduce exposure levels to the new levels.

In reviewing the comments and hearing testimony on the technological feasibility of achieving the PELs and other limits, OSHA has found that for the overwhelming majority of situations where air contaminants are encountered by workers, compliance can be achieved by applying known engineering control methods and work practice improvements. It is recognized, however, that in some circumstances, respiratory protection may be necessary.

Types of Controls

In general, three basic types of controls may be employed to reduce employee exposures:

* Engineering controls * Work practices and administrative reforms * Personal protective equipment Consistent with OSHA regulations and policy, this chapter examines the feasibility of engineering controls and work practices to control employee exposure, in preference to personal protective equipment.

Engineering Controls. Engineering controls involve the use of local exhaust ventilation, general ventilation isolation of the worker and enclosure of the source of emissions process modifications equipment modifications and substitution of non-hazardous chemicals. These methods may be used alone or in combination of any two or more controls depending upon the needs of a specific situation. Variations in situations usually result from the type of process being used and the number of chemicals in the air. However, these controls are considered standard techniques which will effectively control these variables either by themselves, or coupled with changes in work practices.

Ventilation. Perhaps the most widely used technique for controlling chemical exposures is the use of ventilation. General ventilation uses the movement of air within the general work space to displace or dilute the contaminant with fresh outside air. General ventilation is not typically the preferred control method in most operations due to the large volumes of air movement required. Local exhaust ventilation uses much smaller volumes of air, exhausted from the point at which contaminants are generated to remove the contaminant at the source.

Isolation. Isolation involves placing a physical barrier between the hazardous operation and the worker. Many modern, automated manufacturing processes are now fully enclosed in ventilated cabinets. The effectiveness of such a control technique depends on the frequency with which the workers have to enter the enclosure during normal operations. In other situations, rather than placing the process or machine in an enclosure, the worker may be put into a controlled atmosphere enclosure. Many processes which involve potential chemical exposures are operated remotely by operators in air conditioned booths.

Substitution. Substitution refers to the replacement of a toxic chemical in a particular process or work area with another, less toxic product. Properly applied, substitution can be a very effective control technique. However, care must be taken to ensure that the proposed substitute performs in a similar manner to the product being replaced. In addition, it is essential that the substitute be carefully evaluated to ensure that in controlling one hazard, another different hazard is not inadvertently introduced. The substitute must also be compatible with existing manufacturing equipment and processes.

The success of these techniques will depend on the physical properties of the chemicals and emissions encountered (boiling point, vapor pressure, etc.) and the process operating conditions (temperature, pressure, etc.). In some cases, particularly with cleaning solvents, substitution may provide the quickest and most effective means of reducing exposure. In other situations where particular physical or chemical properties are required, major effort may be required to alter processes or install or expand local or general dilution ventilation. The extent to which engineering controls may be effectively used will vary from industry to industry, as well as plant to plant within an industry.

Work Practices and Administrative Reforms. Work practice controls include housekeeping procedures, material handling or transfer procedures, leak detection programs, training and personal hygiene. In many cases, it is possible to bring about substantial reductions in employee exposures by applying work practice controls.

Personal Protective Equipment. Where it is impractical to apply engineering or work practice controls, or where their application will not consistently reduce employee exposures below the final rule PELs, personal protective equipment such as respirators, may be used to prevent and reduce exposures.

Industry Engineering Controls

To determine whether engineering controls and work practices can reduce employee exposures to the final rule PELs, OSHA, through its contractors, examined typical work processes found in a cross section of industries. Using this list, industry experts identified which major processes had potential hazardous exposures and may require additional engineering controls or different work practices in order to achieve the proposed PELs. To assess whether these would be feasible for the processes within the industry group, records maintained by OSHA and NIOSH were searched to identify examples of the successful application of controls to these processes. Based upon the judgments of the industry experts, a determination was made as to the probable feasibility of achieving the proposed PELs. A list of the processes and control measures is set out in Table F-4 at the end of this chapter.

This chapter presents examples of feasible methods of controlling exposure to hazardous substances encountered in processes used in the SICs for which costs and benefits have been identified. Unit costs for these or similar controls were used as the basis for the cost projections in Chapter V. In addition, this chapter summarizes the docket entries regarding technological feasibility.

Information from commenters to the docket or statements at the hearings indicate that for the vast majority of firms, the final rule PELs can be met using engineering controls alone. In the few isolated cases it is recognized that respiratory protection must be added to engineering controls to assure worker safety.

SIC 20 - Food and Kindred Products

A major air contaminant in the food processing industry is carbon dioxide (CO(2)). A milk products plant (SIC 2023) controlled carbon dioxide exposures by using a hood which fully enclosed the chiller-conveyor line and exhausted air from the system to an exterior baghouse. Carbon dioxide levels resulting from the use of dry ice were controlled at a meat packing plant (SICs 2011 and 2013) by a stainless steel exhaust hood. Similarly, a poultry dressing plant controlled carbon dioxide emissions by using a slotted hood exhaust ventilation system. A food processing plant (SIC 202) controlled carbon dioxide exposure by increasing the number of air changes in the packaging room.

OSHA is adopting a limit of 10,000 ppm as an 8-hour TWA for CO(2) and is supplementing this limit with a 15-minute STEL of 30,000 ppm. The Beer Institute [Ex. 49, 142, Tr. 8/9/88, p. 9-26] and the Brewing Industry Safety Advisory Committee submitted comments to OSHA on carbon dioxide. The industry argued at the public hearing and in docket submittals that the 8-hour TWA limit of 5,000 ppm for CO(2) was "unnecessarily low and restrictive" [Ex. 49, Tr. 8/9/88, p. 9-27]. According to the Beer Institute, the brewing industry "is unique relative to carbon dioxide exposure and control....no other industry faces the same engineering difficulties for controlling ambient carbon dioxide as the brewing industry" [Tr. 8/9/88]. No details explaining these difficulties were provided by these commenters.

Monitoring data taken on employees in one brewery, together with a description of the operations that cause the most exposures, are contained in a study of cellar workers [Riley and Bromberger - Barnes, 1979]. The data include samples taken over 14 eight-hour shifts. Eight-hour TWA exposures ranged from 0.5 percent (5,000 ppm) to 1.41 percent (14,100 ppm), with a mean of 1.08 percent (10,800 ppm). Data on "maximum acute exposure" where also provided. The period of maximum acute exposure ranged from 2 minutes to 240 minutes. Of the 14 samples, three exceeded a 3-percent (30,000 ppm) 15-minute STEL.

Exposures result from a build-up of CO(2) in large fermentation tanks during the beer fermentation process. These tanks are sealed systems; the CO(2) is normally piped away. Two circumstances were identified by commenters as causing CO(2) exposures. First, if excessive pressure builds up, an escape valve blows. The concentration of CO(2) in the vicinity of such a blow-out was measured at 60 percent (600,000 ppm), although the level in the area fell to 12 percent (12,000 ppm) within a few minutes. Such blow-outs are reportedly rare. The second, and routine source of CO(2) exposure is the opening of tank doors and the entry of workers into the tank to flush out sludge that remains after the tank has been drained. After opening the doors which are near the floor and open onto the central corridor, the cellar worker leaves the area until most of the CO(2) has been ventilated. The principal exposures to CO(2) in the beer industry thus involve either upset conditions (a blow-out) or maintenance activities (entry into the tank to clean it). For both of these circumstances, OSHA routinely permits the use of respiratory protection. Exposures in the corridor (resulting from the opening of tank doors) could be further controlled by the work practice of cracking the door and waiting longer before reentering the area or by adding local exhaust ventilation to capture the CO(2) escaping from the doors.

OSHA notes that commenters from the brewing industry supported the Agency's proposed STEL for CO(2) of 30,000 ppm [Tr. 8/9/88, p. 9-31], and advocated an 8-hour TWA of 10,000 ppm. In adopting 10,000 ppm as the 8-hour TWA and adding a 15-minute STEL of 30,000 ppm, the Agency believes that feasibility problems in this industry sector will be alleviated.

Grain dust exposures in this sector occur during grain handling operations in facilities that mill grain either for human use, e.g., flour mills and rice mills or, more commonly, for animal use, e.g., feed mills [Ex. 3-752, p. 10]. There is general agreement that the highest exposures in all types of grain-handling facilities occur during grain receiving operations [Ex. 3-752; Tr. 8/10/88, p. 10-46]; the grain receiving process is the same, regardless of the type of facility in which it occurs. OSHA proposed a level of 4 mg/m(3) for grain dust; because of feasibility considerations, the final standard is 10 mg/m(3).

Many commenters stated that a PEL of 4 mg/m(3) was not achievable, particularly in older mills [Exs. 3-63, 3-110, 3-237, 3-299, 3-405, 3-752, 3-755; Tr. 8/10/88, pp. 10-45/10-48; 10-50/10-54; 10-55/10-60; 10-61/10-70]. Industry representatives stated that current employee exposures to grain dust in mills often exceeded the proposed limit [Tr. 8/10/88, pp. 10-63, 10-46; Ex. 180]. For example, David Bossman, representing the American Feed Industry Association (AFIA), reported that "just over half [of 69 samples taken in 10 mills by the AFIA] exceed the proposed PEL. The average exposure was 10.9 milligrams per cubic meter" [Tr. 8/10/88, p. 10-46]. Mr. Bossman also stated that exposures in the bulk receiving areas of all 10 mills sampled exceeded 4 mg/m(3) and averaged 12.9 mg/m(3) [Tr. 8/10/88, p. 10-46].

According to a 1984 study by the T.E. Stivers Organization, 15 of 20 representative mills visited "had no dust control systems at all, and [the remaining] five had some dust control systems, but [these were] not comprehensive in scope" [Tr. 8/10/88, p. 10-63]. According to Gary Winsett, President of Winsett Engineering, Inc., an independent engineering firm that specializes in the feed, grain, and related agribusiness industries: "three separate control systems would be required in the main work areas of each mill" to bring 13 of the 20 mills included in this 20-mill survey down to the 4 mg/m(3) level of control, and six of these 13 mills would require "relatively extensive dust control systems in the receiving areas" to achieve the 4 mg/m(3) limit [Tr. 8/10/88, p. 10-63]. In cases where such controls are in place, however, Mr. Winsett reported that exposures had been reduced considerably [Tr. 8/10/88, p. 10-62]. In older mills, retrofitting has been successful in reducing grain dust exposure levels. For example, John Wolgemuth, Corporate Safety and Loss Control Manager for Agway, a farm supply and food marketing cooperative owned by 102,000 farmer-members, described the results achieved in one mill in which additional exhaust hoses had been installed. According to Mr. Wolgemuth, levels were reduced from above 15 mg/m(3) to below the 10 mg/m(3) level by retrofitting [Tr. 8/10/88, pp. 10-50/10-51].

In an effort to obtain information on conditions in small, rural mills, OSHA reviewed the docket developed in connection with the Agency's recent grain-handling standard [Docket H-0117]. A study performed by Dr. Buchan of Colorado State University reported that, in eight small grain elevators and feed mills in his state, 10 percent of exposure samples were above the 4 mg/m(3) proposed limit (Attachment 1, Ex. 3-751, Docket H-0117).

There are a variety of dust controls in use in grain mills at the present time. Dust collection systems, including pneumatic dust controls, are the most widely available and useful methods of controlling grain dust in mills in which dust is a problem [Ex. 3-752, p. 17; Tr. 8/10/88, p. 10-62]. A dust collector typically consists of a motor-driven fan, which creates the air flow necessary to capture dust particles and carry them through duct work to a dust collector. These aspiration systems are an "effective method of controlling dust emissions. Aspiration of the leg consists of a flow of air across the entire boot, which entrains the liberated dust and carries it up the up-leg to take-off points" (52 FR 49592, December 31, 1987). Depending on baseline levels of exposure, several collectors may be needed in a mill.

A second method of controlling dust that is becoming widely used is the application of oil mist to the grain to minimize dust generation. This oil mist, which consists of mineral oil, vegetable oil or some combination of the two, is normally applied when the grain is received at the mill. Ralph Mourer, testifying for the AFIA, stated that oil suppression of dust is a promising control that he has just installed in his feed mills. Although he has not yet had much experience with the system, he noted "people I've talked [to] and discussed the system with are very pleased" [Tr. 8/10/88, p. 10-78]. In an earlier study of grain handling facilities for OSHA, however, Arthur D. Little Inc. noted that there are some limitations to the use of this method:

Mineral oil is not approved for use as an additive on food grades of grain by the U.S. Food and Drug Administration. Vegetable oil may be an allowed additive, but its use can cause the grain to adhere into masses in cold climates. Further, there is concern that the oil will become rancid or create a commercially objectionable odor (Docket H-0117, ADL, p. VI-34).

Scott Bjornsom from Hunter Grain in North Dakota also reported that oil suppression cannot be used for malting barley "because of the absorption with the water in the malt process" [Tr. 8/10/88, p. 10-85]. Despite some limitations on its use, oil suppression appears to be an effective control.

A third control method that can be used in facilities with high dust exposures is the use of vacuum systems in place of manual sweeping or compressed air blowing during clean-up operations. A Canadian study [Farant and Moore, "Dust Exposures in the Canadian Grain Industry," American Industrial Hygiene Journal, March 1978] of grain elevators found that many very high exposures to grain dust were a result of dust raised during housekeeping operations that involved brooms or blowers [Page 193, Attachment to Ex. 3-751, Docket H-0117]; this study concluded that "the use of in-plant vacuum systems would reduce these exposures." Representatives of the AIFA reported that some mills have switched to vacuum systems to control their employees' exposures during clean-up [Tr. 8/10/88, p. 10-78].

Industry representatives also described the filtration systems that are being installed on aspiration systems in feed mills; these systems re being installed in many areas of mills, especially in the loading and unloading areas, where the highest exposures to grain dust occur [Tr. 8/10/88, p. 10-73]. In the past, unfiltered cyclones were in widespread use in feed mills [Tr. 8/10/88, p. 10-84].

To deal with the problem of grain dust in older mills, many owners are replacing the old-fashioned wooden legs with "good, tight, enclosed steel legs ....[and in] old facilities...[that had] open grain drag conveyors...the conveying systems that used to be open have lids on them...to keep the dust where it belongs" [Tr. 8/10/88, p. 10-80]. Enclosure of this type is a standard and recommended industrial hygiene practice in all dusty environments.

OSHA notes that much of the control and exposure data relied on by AFIA representatives at the hearing, such as the 1984 Stivers study of the representative group of 20 feed mills, predates the promulgation of OSHA's grain handling facility standard; the Agency believes that many facilities in this sector are in the process of replacing outdated equipment, retrofitting existing equipment, and "tightening up all connections" throughout the mill [Tr. 8/10/88, pp. 10-82/10-83]. This is confirmed by industry representatives, who reported that these efforts are being undertaken in response to the OSHA standard, their insurance companies' suggestions, and industry concerns about dust levels [Tr. 8/10/88, p. 10-73]; according to industry representatives, these controls have reduced fire risks [Tr. 8/10/88, p. 10-74], improved productivity and quality, and led to better working conditions [Tr. 8/10/88, p. 10-81].

OSHA's review of all of the evidence in the record indicates that 10 mg/m(3) is a feasible limit in the grain and feed mill sector. The final rule includes this PEL as an 8-hour TWA; the Agency finds that the health evidence (see Section XIC of the preamble to the final rule) demonstrates a significant risk of material health impairment above this level. OSHA finds that feed mill employers will be able to achieve the 10 mg/m(3) limit in cases where exposures remain above 10 mg/(m)3 [see Tr. 8/10/88, p. 10-46] or where older mills are involved [see Tr. 8/10/88, p. 10-63] using any of a variety of controls: oil mist suppression in feed mills, aspiration systems (with or without filtration), enclosure of open conveyors and other grain-handling equipment, and the use of vacuuming in lieu of blowing or sweeping during cleanup. For some mills that are close to this limit at the present time, OSHA believes that the general "tightening up" described by Mr. Wohlgemuth [Tr. 8/10/88, pp. 10-82 to 10-83] will be sufficient.

The International Institute of Ammonia Refrigeration (IIAR) argued that the proposed levels for ammonia (25 ppm TWA, 35 ppm STEL) would be viewed as a nuisance because most people cannot detect the odor of ammonia at 35 ppm. As such, employees would neglect proper control measures [Ex. 113]. David G. Kramer of Kahn's and Company [Ex. 113] stated that "No one in our plants is exposed to continuous exposure to 35 ppm ammonia concentrations".

One control approach for ammonia gas encountered in poultry processing (SICs 2016 and 2017) required the appropriate placement of cut-off valves to freezer coils and the use of an alarm detection system to monitor ambient air conditions.

Ammonia-based refrigeration systems are commonly used in the meat products industry. Commenters expressed concern that "ammonia based refrigeration systems...are subject to occasional leaks which may result in short-term high level exposures" [Exs. 3-897, 3-750]. The situation referred to by these commenters is an intermittent maintenance or upset condition, for which OSHA permits the use of respirators. In addition, a representative of the Food and Allied Services Trade Department of the AFL-CIO stated that two companies, Wilson Foods and Morrell, evacuate the workplace if ammonia levels reach 25 ppm as a ceiling [Tr. 8/4/88, p. 311]. OSHA concludes that there is no issue of technical feasibility in regard to the proposed STEL of 35 ppm for ammonia and the 35 ppm STEL is retained in the final rule.

Chlorine is used extensively as an antibacterial agent in meat products plants to comply with USDA sanitation and microbiological contamination requirements. Commenters did not raise the issue of technical feasibility in regard to the proposed chlorine standard itself. Commenters did, however, express concern that a 0.5 ppm STEL for chlorine may be too stringent to allow compliance with USDA regulations [Exs. 3-756, 3-897], although no data to support this concern were provided. Responding to these concerns, OSHA has established a 0.5 ppm PEL and 1 ppm STEL for chlorine in the final rule.

Carbon disulfide itself is not used in the meat products industry, although it is a key solvent used in the manufacture of cellulosic food casings, which are used in the manufacture of processed meats. Suppliers of cellulosic food casings stated that a carbon disulfide standard of 1 ppm cannot be met in the production of such casings [Exs. 3-421, 3-633, and 3-897]; if this were the case, according to these commenters, domestic supplies of cellulosic casings would cease. Foreign supplies would gradually penetrate and supply the market for cellulosic food casings [Tr. 8/2/88, pp. 4-209, and 4-261]. OSHA concludes that there is no apparent issue of technical feasibility of the proposed carbon disulfide standard in SIC 20. However, the TWA for this substance has been increased to 4 ppm, in part, in consideration of the potential industrial displacement effect.

The National Cotton Council of America (NCCA) submitted a comment to the effect that the approximately 50 cotton mills in SIC 2074 would be adversely affected by the proposed limit for n-hexane and other hexane isomers, vegetable oil mist, and grain dust [Ex. 3-1080]. NCCA stated that its members would have difficulty measuring airborne concentrations of these substances because cottonseed mills are small, rural business without in-house industrial hygiene capability. OSHA notes, however, that methods are readily available to measure these airborne contaminants; an appendix to the final rule contains information on appropriate sampling methods for these substances. The Agency has responded to industry concerns by dropping its proposed 10 mg/m(3) STEL for oil mist but retaining the 5 mg/m(3) TWA.

Sulfur dioxide (SO(2)) exposures in the wet corn milling industry as a result from soaking of cleaned corn kernels in large vats (known as steep vats) for 30 to 50 hours. The purpose of the steeping process is to soften the corn in preparation for further grinding, screening, and centrifugal operations. This steeping process takes place in a dilute sulfur dioxide solution (sulfurous acid) [Tr. 8/15/88, pp. 9-10]. Worker exposures occur when sulfur dioxide is released from solution in the steeping tanks. The principal controls available to reduce exposures to sulfur dioxide are local exhaust ventilation, the use of isolated control rooms, process enclosure achieved by the use of closed stainless steel tanks, enclosed screening systems, and general automation [Tr. 8/15/88, pp. 8-77 to 8-78].

Exposure data for this segment of SIC 20 are sparse, except for data from a study conducted by the CRA in five of its member plants in 1977 in connection with the Agency's earlier SO(2) rulemaking. Eight-hour TWA samples were taken on 213 workers exposed to SO(2) in wet corn milling and on a group of 344 non-SO(2)-exposed workers from other parts of the plant [Ex. 65, Tab 9]. (The "background" SO(2) level even for the controls, however, was determined to be 0.33 ppm (8-hour TWA).) The median exposure in the SO(2) group was 2 ppm; 15 percent of all workers were exposed above 5 ppm. No STEL measurements were taken. Exposures (8-hour TWAs) ranged from 0.1 to 10.8 ppm. According to industry sources, these results "represent worst-case" exposures because they were obtained during the winter months, when the plants' windows and doors were closed [Ex. 65, Tab 9, p. 7]. More recent exposure data, described at the hearing as "non-systematic" and variable in "sampling efforts, methods and results," were summarized as follows:

While many plants report 1987-88 personal sampling results in the range of 2 ppm, even plants in that category are not below 2 ppm consistently, and a large number of employees are still exposed in the range of 4 to 5 ppm.

Industry representatives at the hearing indicated that opening the doors and windows even when there was only a 5-mph breeze outside increased the effectiveness of in-plant ventilation by a factor of five [Ex. 65, Attach. D, pp. 2392-2397]. In addition, testimony indicates that the higher 8-hour TWA readings and those above 10 ppm were caused by "emergencies," "pipes breaking, or the process getting out of control, the tank...[overflowing] as a pump seal breaks, or something of that sort" [Ex. 65, Attach. D, pp. 2314, 2315, 2319]. Testimony also indicates that many of the sampling results reported above were taken on maintenance workers, who are personnel dedicated to maintenance functions [Tr. 8/8/88, p. 8-85].

Industry representatives reported that major improvements in SO(2) exposures could be achieved:

Plants vary widely in age, the degree of natural ventilation available, the degree to which their process is entirely closed, the location and source of the sulfur dioxide they use in steeping, the amount of local exhaust equipment already in place, the extent to which control rooms isolate the operator from the process, and various other factors [Ex. 65, Attach. F, pp. 35, 37].

Some equipment, such as the "steeps" or soaking tanks, are more than 40 years old; some of these tanks are still the wooden staved steeps of years ago [Ex. 65, Attach. D, p. 2324]. One company has milling plants that range from 30 to 97 years in age [Ex. 65, Attach. D, p. 2324]. Spokesmen reported that the industry's efforts to modernize plants has not resulted in appreciably lower employee SO(2) exposures because improvements in engineering controls, i.e., ventilation, have not kept up with increased production [Ex. 65, Tab 13, p. 7]. An OSHA-sponsored study performed by JRB Associates for the previous SO(2) rulemaking found that plants in this sector could achieve the 2-ppm TWA and 5-ppm STEL with the expenditure of a relatively small amount of money [Ex. 65, Attach. D, pp. 2322-2324].

There are no sampling data in the record relating to the 5-ppm STEL for SO(2), because the CRA-sponsored exposure survey undertaken in 1977 contained no STEL sampling results. The recent record [Ex. 65, Tab 13, p. 7] states simply that:

Short term exposures, especially for maintenance job functions, can be considerably higher than 4 to 5 ppm [Ex. 65, Tab 13, p. 7].

OSHA notes, however that the wet corn milling process is a steady-state process:

The process...is rather level as far as the [SO(2)] concentration is concerned with the exception of emergencies, pipes breaking, or the process getting out of control....(Ex. 65, Attach. D, p. 2314-2315)....[except] for maintenance emergency problems, the exposure to sulfur dioxide in the process is fairly constant.

OSHA finds that the 2-ppm 8-hour TWA and the 5 ppm 15-minute STEL for sulfur dioxide are technologically feasible in the wet corn-milling process. The Agency's reasoning is as follows:

(1) In 1977, 50 percent of all SO(2)-exposed employees had exposures at or below 2 ppm; because the sampling results for dedicated maintenance employees are contained in these numbers, the actual percentage is greater than 50 percent for non-maintenance workers;

(2) Most of the sampling results from the high end of the 0.8 to 10.8 ppm range of exposures cited by the CRA occurred during process upset or maintenance operations;

(3) The 1977 CRA sampling results were "worst case," so the number of overexposed employees is overstated; respirators are permitted in these operations;

(4) Because most exposures are already below 2 ppm, little in the way of additional control will be needed (note that opening the windows increased the efficiency of ventilation by a factor of 5, indicating that additional make-up air would do the same);

(5) STEL exposures are not a problem because the wet milling process, except when it is not being adequately controlled, is characterized, according to industry representatives, by relatively constant, non-fluctuating ambient concentrations of SO(2). Because most exposures are already below 2 ppm and the overwhelming majority are now at or below 5 ppm, the STEL has essentially been achieved in this sector. That is, in a steady-state exposure environment where 8-hour TWA exposures are below 5 ppm, 5 ppm STEL exposures are not a problem. In wet corn milling, short-term exposures are a problem only in maintenance and emergency operations; in both cases, respirators are both permitted and encouraged by OSHA.

(6) Where exposures are above 2ppm, they are only slightly above 2ppm. Minor upgrades in ventilation and some modernization of the oldest equipment will reduce exposures below 2 ppm.

SIC 21 - Tobacco Products

Tobacco dust and residual pesticide dusts created during cutting and shredding operations have been reduced through the use of local exhaust ventilation. This has also been used to control emissions of ethyl alcohol-based chemical flavorings during blending operations.

There were no comments submitted to the docket for this sector.

SIC 22 - Textile Mill Products

Textiles are dyed at various stages in their manufacture, including unspun fibers, unwoven yarn, and finished fabric. Workers who prepare fabrics from unspun fibers are of particular concern, since they could be potentially exposed to dyes contained on dusts generated during manufacture. In addition, some dyes possess much poorer fastness to wet treatment than do others; persons who launder such clothing are potentially exposed to the dyes. Stringent control measures and work practices can prevent such exposure.

Several generally acceptable practices for the control of hazardous materials can be used wherever there is the potential for exposure. For example, pressure failure alarms for closed systems and exhaust ventilation can rapidly indicate a system failure that might result in the release of substantial quantities of dyes. Continuous flow indicators, such as water or oil manometers properly mounted at the juncture of a fume hood and duct throat and marked to indicate acceptable airflow, will give a readily observable indication of decreased efficiency in the ventilation system for the hood. Wet methods, vacuum cleaning, or other methods that do not lead to redispersion of settled dust should be used for plant maintenance and sanitation. Dry sweeping or blowing with compressed air should be prohibited.

Evidence presented in the docket suggest that controls necessary to meet the proposed standards have already been installed at many facilities. The American Textile Manufacturers Institute, Inc. (ATMI), representing 85 percent of the industry's manufacturing capacity, reported that "member companies generally try to meet the ACGIH TLVs for both those chemicals which are regulated by PELs and those which are not. Because the ACGIH TLVs are annually reviewed and revised, ATMI's member companies believe compliance with these voluntary standards has led to safer and healthier workplaces for their employees" [Ex. 3-434].

The National Cotton Council of America [Ex. 3-1080] reported that "Textile manufacturers generally try to use the existing ACGIH TLVs as guidelines for good practice to provide a safer and healthier workplace for their employees." Their comments state that some of the levels are difficult to attain, but are said to be feasible.

SIC 23 - Apparel

Chemical exposures in the apparel industry occur principally as a result of three exposure sources: spot cleaning, dry cleaning and contact with treated fabrics.

Spot cleaning and dry cleaning operations resulting in exposures to perchloroethylene can be controlled with the use of local exhaust ventilation and general ventilation. Work practice improvements help reduce solvent exposure during transfer operations. Routine scheduled maintenance can be used to detect and control leaks from door gaskets and seals. Contact dermatitis is reduced through the use of disposable gloves and adherence to a personal hygiene program.

A detailed discussion on perchloroethylene is presented in SIC 72.

SIC 24 - Lumber and Wood

The primary worker exposure in the lumber and wood industry is wood dust. For the operation of large equipment (e.g. in debarking and sawmill activities), the operator can be placed in an enclosed control booth, or in the case of moving equipment (e.g. cherry pickers, loaders and cranes), the operator can be located in an enclosed cab. In both cases, air would be filtered and conditioned. In the case of felting or matting process lines, or such equipment as belt sanders, the equipment can be enclosed or hooded and vented to a baghouse. For smaller equipment, such as variety saws, tenoners, and dovetailers, hoods or various types of negative pressure (or combinations of positive and negative pressure) local ventilation devices can be used to control wood dust. In the case of hand-held sanders, a vacuum system can sometimes be applied to the process. Some other wood dust generating equipment can also be enclosed (e.g. planers), but this is generally done for noise control.

The technical feasibility of a 5 mg/m(3) PEL for wood dust was challenged, indirectly, by only one commenter to the record. The American Furniture Manufacturers Association [Ex. 3-917], after speaking of the general technical feasibility of the proposed standard and the difficulty of controlling wood dust around some machines, stated: "Other machines are so complicated (such as multiple spindle boring machines and multiple spindle carvers) that no effective collection system has yet been defined." OSHA disagrees and concludes that exposures on these machines can be controlled. Included in the documentation of the site visits conducted for this rulemaking [Ex. 11] is at least one case of a multiple head boring machine which was equipped with local exhaust ventilation and a multiple spindle "trim, bore and dowel" machine also equipped with local exhaust ventilation. TWA exposures to wood dust for the operators of these machines were 1.0 and 0.4 mg/m(3), respectively.

Vast numbers of commenters expressed their support for a 5 mg/m(3) PEL for wood dust without any question of technological feasibility. A few examples follow. Appalachian Hardwood Manufacturers, Inc. [Ex. 3-626] stated that, although they felt it would be expensive, "To bring all our mills into compliance with a five milligram per cubic meter standard would be technically feasible." Monadnock Forest Products, Inc. [Ex. 127, Attachment C and Tr. 8/12/88, p. 216] states that "5 mg/m(3) is technically feasible but due to cost it should be phased in over a number of years." The National Dimension Manufacturers Association [Ex. 3-1160] commented: "Achievement of a 5 mg/m(3) permissible exposure limit [for wood dust] is believed to be technically feasible...." Others at the hearings supporting the adoption of the 5 mg/m(3) level included David Smith of Willamette Industries [Tr. 8/12/88, p. 369] and Charles Carey of Ross Associates [Tr. 8/12/88, p. 411]. Whirlpool Corporation [Ex. 3-824] provided exposure data for a sanding work station, before and after the installation of control equipment. Exposures before the ventilation equipment was installed ranged from 13.0 to 29.6 mg/m(3). With the equipment in place, exposures ranged from 0.88 to 3.16 mg/m(3).

Two surveys cited by Mr. Scott Schneider of the Workers' Institute for Safety and Health [Tr. 8/15/88, p. 6 and Ex. 115, Attachment A] also support the feasibility of the 5 mg/m(3) PEL. A 1986 OSHA Health Response Team survey showed that "two thirds of the personal samples were below two milligrams per cubic meter and over 40 percent were below one." Twelve of the 23 plants in a 1985 survey by Haliday Associates in Ontario, Canada, had no exposures over five milligrams per cubic meter and two of the plants, one of which was a furniture plant, had no exposures above one milligram per cubic meter.

Exposure data from the Clayton Environmental Consultants' study for the Inter-Industry Wood Dust Coordinating Committee was cited by Mr. Michael Coffman at the informal public hearings [Tr. 8/12/88, p. 99]. Mr. Coffman stated: "Within SIC Code 24, we collected a total of 676 dust measurements. Eight percent of these were found to exceed five milligrams per cubic meter; 30 percent exceeded one milligram per cubic meter. Within SIC Code 25, 107 total dust measurements were collected. Fifteen percent of these exceeded five milligrams per cubic meter, 40 percent exceeding one milligram per cubic meter. Within SIC Code 26, a total of 19 measurements were collected, five percent exceeding both one and five milligrams per cubic meter."

Exposure data presented by machine type in the Clayton study [Ex. 127A], and shown below in Table F-1, provide clear evidence that the 5 mg/m(3) level can be attained for many machines by using local exhaust ventilation. These data demonstrate that exposures of operators at these machines can be uniformly controlled. Table F-2 presents data showing the effectiveness of ventilation and air-conditioned booths for other machines.


                   Table F-1

           EXPOSURES TO WOOD DUST FOR MACHINES
         EQUIPPED WITH LOCAL EXHAUST VENTILATION

                      Number of Exposure Measurements
                      _______________________________

Machine                   Total      Above 5 mg/m(3)
_______                   _____      _______________
Bandsaw-finish                7                    1
Cut-off saw                  33                    1
Gang rip saw                 15                    2
Rip saw                      22                    0
Variety saw                  15                    1
Belt sander                  41                    3
Hand-held sanders             8
Molder                       17                    0
Planer                       16                    2
Router                        4                    0
Shaper                        9                    1
Tenoner                      22                    5


                               Table F-2

                  EXPOSURES TO WOOD DUST FOR MACHINES
             WITHOUT CONTROLS OR WHERE OPERATOR IS ISOLATED
                        IN AIR-CONDITIONED BOOTH

                        Number of Exposure Measurements
                        _______________________________
                                   Above
Machine                Total     5 mg/m(3)   Controls
_______                _____     _________   ________
Drilling and Boring        7         1         No controls
Sorter/Stacker            12         1         No controls
Chipper                   12         1         No controls
Dryer                      7         0         No controls
Veneer Clipper             4         0         No controls
Hot Press                  5         0         A/C booth
End Loader                 6         0         A/C booth
Felter                     3         0         A/C booth
                           5         0         Ventilation
                           4         1         No controls
Bandsaw-Sawmill            8         0         A/C booth
                           4         1         Ventilators
                          10         2         No controls

A number of commenters also estimated what their costs of compliance would be if a 5 mg/m(3) standard were adopted (specific examples are addressed in Chapter G). Estimating the costs implies the technical feasibility of meeting the standard. All of the foregoing is evidence that control of wood dust at or below 5 mg/m(3) is technologically feasible.

Washington State has adopted exposure limits of 2.5 mg/m(3) for western red cedar and 5 mg/m(3) for other woods. According to Mr. Stephen Cant of the Occupational Health Program for the State of Washington, "...in our process of adopting this specific wood dust standard the industry presented absolutely no comment in terms of concerns regarding the limits" [Tr. 7/29/88, p. 102]. Mr. Cant also states: "...Industry has been comfortable with [these] limits in the northwest. We find that they can, in fact, in most cases, comply with those limits,...."

A study conducted by the University of Washington Department of Environmental Health [Ex. 127H] provides recommendations on achieving the 2.5 mg per cubic meter level, as well as some exposure data. The report notes that average exposures for shake saw operators in the mills surveyed were 1.63 mg/m (3). The range of exposures for other mill workers (splittermen, deck hands and packers) ranged from 0.22 to 2.7 mg/m(3). Shingle saw operators were the only workers routinely exposed at levels above the 2.5 mg/m(3) limit, with average exposures of 3.84 mg/m(3). From this study: "The Washington exposure levels can be compared to levels in Canadian sawmills after ten years of a Canadian regulatory limit of 2.5 mg/m(3). Cedar dust levels were studied in Canadian sawmills in 1985 by a researcher named Vedal. Total dust exposure in observed Canadian mills ranged from undetectable to 6.0 mg/m(3) averaged over an 8-hour work day, with an average dust exposure of 0.21 mg/m(3). Only 10% of workers were exposed to more than 1.0 mg/m(3), and of these only 3.9% had exposures greater than 2.0 mg/m(3) TWA." Included in the University of Washington report are a number of specific designs for local ventilation, baffles for shake and shingle saws and general recommendations on housecleaning.

SIC 25 - Furniture

The feasibility of the standard for wood dust in this SIC is discussed in conjunction with SIC 24, above.

A review of processes in the metal office furniture manufacturing sector (SIC 2522), shows that air contaminants from the coating process have been controlled. Prefabricated sheet components for file cabinets are prewashed and coated with polyester or acrylic on a high speed conveyor line. The application process includes manual spraying of cabinets with airless atomizing sprayers and electrostatic spray guns on reciprocators. Manual spraying operations are performed in downdraft booths. Filtered fresh air is supplied through the open top of the booths and removed at the bottom through a water curtain by exhaust fans mounted on the roof of the booth. Spray headers in exhaust plenums clear paint mist from the air stream. Automatic spray booths contain electrostatic spray guns and side draft ventilation. Furthermore, organic solvent vapors in the paint mixing and storage room are controlled by equipping each drum with a heavy barrel cover, an integral agitator, sealed pipe openings, and a closeable access line.

Masco Corporation [Ex. 3-682] stated: "Methodologies for control of solvents from finishes in the woodworking industry are limited....OSHA therefore has presented no feasible methodology for the woodworking component of the furniture industry to control solvent or potential gaseous air toxics." OSHA concludes that control is feasible. The control for such potential exposures is, in almost all applications of lacquers, varnishes, sealers and stains, a form of local exhaust ventilation control commonly referred to as a spray booth.

Spray booths are in wide use in the furniture industry and were seen in use at several of the plants on the site visits conducted for this rulemaking analysis [Ex. 11]. In each case the exposures were far below the PELs, and in several cases the solvent levels were not detectable.

A NIOSH study [TA 79-047-825] recommends that the exhaust opening in a painting room should be located as close to the painting operation as possible to take advantage of spot ventilation. The exhaust opening location should be such that overspray is directed into the opening. The exhaust outlet and air inlet should be placed so that all the air used for ventilation passes through the zone of contamination. The air flow should direct all contaminants away from the painter's breathing zone and into the exhaust outlet.

Most welding in furniture manufacturing occurs in a fixed location where exposures to the various components of welding fumes can best be controlled with adequate local ventilation. Numerous manufacturers have available local exhaust systems and source collection/filtration systems that will control welding fumes. These systems typically consist of a fan and either a fixed hood or a jointed, flexible arm, up to fifteen feet in length, at the end of which is a small hood. The flexible arm allows the exhaust system to cover a large work area. Such systems operate at air volumes of 400 to 1000 cfm and can be exhausted to an existing centralized ventilation system, to the vicinity of general shop ventilation (e.g., a roof fan) or directly to the outside. A wide variety of off-the-shelf local exhaust systems and collection filtration systems are available, including portable models. Custom ventilation systems can also be installed.

SIC 26 - Paper and Allied Products

Pulp mills occur primarily in SIC 2611, but can also be present as part of the operations in SICs 2621 and 2631. High control costs could potentially be incurred because of the large quantities of chemicals used in breaking down pulp to form cellulose and the reactions that occur in the digesting process between these chemicals and the substances contained in the wood fiber. Large quantities of chemicals such as chlorine and sodium hydroxide are also used in the bleaching operations. The digesting and bleaching operations are also very extensive. Large quantities of pulp are generally produced from wood in these mills either for captive use or for shipment to paper and paperboard mills. The type of controls that would be used include ventilation, enclosure and/or process change, but less likely the latter. Various engineering controls have been used by the paper mill industry to prevent the mixing of toxic chemicals in sewer lines. Tanks containing the hazardous chemicals have been isolated and surrounded by dikes. Discharge lines have been re-routed to prevent accidental mixing. OSHA believes that feasible controls are available.

Wood dust can be generated in some processes in pulp mills in SICs 2611, 2621, and 2631. Workers may be exposed to wood dust from debarking and chipping operations, as well as in the wood yard. Exposures to equipment operators in the wood yard can be controlled by installing enclosed, air-conditioned cabs on the equipment. Debarker operators are frequently protected by isolation in air-conditioned booths. Exposure data for debarkers included in the Clayton study of the Inter-Industry Wood Dust Coordinating Committee show only 2 of 21 measurements above the 5 mg/m(3) level. Data for chippers, also from the Clayton study, suggest that controls are rarely needed for chippers: Only one of the 13 exposure measurements for chippers that had no controls in place was in excess of 5 mg/m(3). Based on all of the above, OSHA concludes that controls are feasible for this industry.

SIC 27 - Printing and Publishing

The technological feasibility for the proposed standards for toluene was challenged by R.R. Donnelley and Sons (Ex. 3-916). "Donnelley believes that compliance with the proposed 150 ppm ceiling will be infeasible during certain press operations and especially during cleaning." Donnelley further states that "if the 100 ppm standard is achieved at great cost through general ventilation improvements, compliance with the 150 ppm 15-minute ceiling would be impossible without the use of local ventilation or respirators." The main concern of this discussion becomes apparent by the following remarks. "If, as proposed by the regulation, respirators cannot be used six months after the effective date of the regulation, local ventilation is the only option. Given the number of presses in service at Donnelley's plants...the proposed 150 ppm ceiling will require, at a minimum, a maze of ventilation equipment which can be expected to cost millions of dollars." These statements reflect the costliness of achieving the final rule standards for toluene in this firm but acknowledge that engineering controls are feasible. The commenter apparently misunderstood the discussion (Federal Register, Vol. 3, 6/7/88, pp. 21241,2) on the proposed length of time for firms to achieve compliance under the hierarchy of controls. OSHA also believes that there are substitute solvents available such as ethanol, ethyl acetate or nitroethane that could be used in cleaning operations or water-based inks not requiring major solvent use that could enable printing firms to achieve compliance with the final PELs. In addition, OSHA believes that engineering controls in the form of local exhaust ventilation are technologically feasible for press applications in the printing industry.

Site visits and monitoring were conducted to two printing establishments [Ex. 11 - Firms 14, 45). One was a letterpress and lithographic operation and the other was a "quick-print" shop. In neither case were there overexposures to any of the following chemicals in use: antimony and compounds, dipropylene glycol methyl ether, ethylene glycol vapor, hexane isomers (other than n-hexane), n-hexane naphtha, potassium hydroxide, propylene glycol monomethyl ether, sodium hydroxide, stoddard solvent (mineral spirits), tetrachloroethylene, and inorganic tin compounds (except oxides).

SIC 28 - Chemicals and Allied Products

In its research on technological feasibility, OSHA found the following examples of controls currently in use:

* The plastic materials and resins manufacturing sector (SIC 2821) used a tank with a hinged cover and fixed ductwork as an exhaust when dumping dye and additives into hot methanol.

* Dust exposure during the bag opening operation in paint manufacturing (SIC 2851) was controlled by modifying the hood to increase dust capture. Likewise, a new dust collection system (collection hoods) with increased capture velocity was installed for use in the bagging and packaging of pesticides (SIC 287).

* Pharmaceutical manufacturers (SIC 2824) addressed the problem of nuisance dust particles by fitting vacuum crescents and elephant trunks on point sources, by fitting chutes with covers, and by placing vacuum attachments on receiving drum covers. Additionally, monitoring was performed from an outside room.

* In order to reduce employee exposure to sulfur dioxide while producing sulfur dioxide gas (SIC 2819), sample collection units were enclosed and attached to a fume collection system. Sample waste was recycled to prevent open exposure in process areas. Electronic spent acid interface detectors were installed to eliminate the need for employee visual inspection of intermittently pulled samples.

* To control TDI exposure in urethane foam manufacturing (SIC 2822), the bun conveyor was enclosed and exhausted. Employee exposure was limited to the startup and finish procedures when installing and removing bun support. A mechanism was designed to support the bun, which eliminated the need for it to be done manually.

In addition, OSHA looked at controls used in paint manufacturing processes. The production of paints (SIC 2851) is a batch procedure which involves the following steps: prebatching, mixing, dispersing, tinting and shading, filling, and storage or shipping. When prebatching or mixing, an employee will slit a bag of dry pigment with a knife and either scoop out the contents for weighing or dump the pigment into the mixer. In some cases, pigments are received in a slurry form and are piped directly into the mixer. Solvents and other raw materials are added into the mixer. Once combined, the mixture is in a paste or slurry form. This mixture is then thoroughly dispersed in a roller, ball, or sand mill or a high-speed disperser all of which are generally closed processes. The paste is transferred to a storage tank where thinning or other agents are added. The paint is later drawn off, filtered and packaged in cans or drums. Airborne dust exposures to components in dry pigments occur during the prebatching and mixing operations when the bags of pigments are opened and dumped. Exposure to chemicals in dry pigments can also occur from pigment spillage and empty bag flattening and disposal. Once the batch is in solution in the mixer, there are no further dust emission points. Exposure to solvents can occur during addition of these ingredients to the mixing tanks, during any leaks or spills, and during packaging.

Local exhaust ventilation would be used to control exposures to dusts and fumes in the paint production processes. Pigment dust exposures at the dumping station can be controlled with the use of a vented enclosure kept under negative pressure by a ventilation system. Empty bags would be manually ejected through a side opening into a large plastic disposal bag to minimize dust generation during bag flattening and disposal.

Exposures to solvents would be minimized with the use of portable hoods attached to flexible ductwork. These ventilation hoods could be placed over the liquid dumping process and also the packaging operation if the percentage and volatility of the solvents would result in exposures.

Observations and judgments proffered by various chemical industry representatives and associations indicate general compliance with the PELs. Such statements indicating widespread compliance demonstrate the existence of available and practicable control methods for a number of chemicals and processes.

Technological feasibility in SIC 28 for most of the proposed PELs is not challenged in the record. Comments received from ARCO [Ex. 3-740] state: "In general, the petrochemical industry has been using the ACGIH TLVs as the primary workplace exposure guideline for years." This statement implies that most of the PELs are not only feasible but are currently being met.

Dr. Isadore Rosenthal has stated on behalf of Rohm and Haas Company that "experience and data tell us that it is feasible for our company to achieve the ACGIH TLV workplace exposures. This takes time to accomplish, however, and therefore a phased in approach to controls is necessary. The period of time in which a firm has to achieve exposure controls should begin only after OSHA has certified a feasible analytical method" for determination of exposure [Tr. 8/1/88, pp. 15-16].

The Polyurethane Manufacturers Association (PMA), in expressing support for the proposed PEL for 4,4'-methylene bis(2-chloroaniline), also discussed the feasibility of achieving the proposed level of control [Ex. 3-683]. PMA stated that "...the various control technologies and personal protective equipment for these various situations [where exposures occur] is recognized in the industry...". Representatives of the PMA also testified that they believe the industry can comply [Tr. 8/9/88, pp. 83 and 91].

Feasibility of controlling exposures to talc dust was indicated by the remarks of the R.T. Vanderbilt Company [Ex. 3-108]: "We would agree with the ACGIH that dust control has all but eliminated the excess death rates in the talc industry. We also support the 2 mg/m(3) respirable talc dust standard." Vanderbilt apparently foresees no difficulty in controlling talc dust at the new PEL.

The feasibility of the proposed 50 ppm PEL for styrene was asserted by the Dow Chemical Company in its comments to the record [Ex. 3-741]: "Dow manufactures styrene and uses styrene in several processes including the manufacture of styrene polymers and polyester resins. These operations can be controlled to reduce exposures below the proposed PEL of 50 ppm and, in fact, most Dow operations already operate at less than 50 ppm."

The Halogenated Solvents Industry Alliance (HSIA) expressed some concern about the feasibility of the proposed 2 ppm 60-minute STEL for carbon tetrachloride [Exs. 3-873, 8-86, 186], but did not identify specific areas where compliance might be infeasible. HSIA also stated "Since carbon tetrachloride is generally used as a process solvent or raw material, workplace exposures are quite low, generally, we believe, below the ACGIH TLV of 5 ppm." Dow Chemical identified specific tasks and operations such as sampling, loading and unloading, and maintenance where they felt that compliance might be difficult or impossible [Ex. 3-741].

Similar concerns were raised about the feasibility of the proposed 2 ppm STEL for chloroform. HSIA anticipated "that non-chemical industry users...would find a 2-ppm limit infeasible in some cases" [Ex. 3-873, 8-86, 186]. The only tasks or processes specifically mentioned in the record as potential problem areas for carbon tetrachloride and chloroform are sampling, loading and unloading, and maintenance [Dow Chemical Company, Ex. 3-741]. These three tasks were named by Dow as problem areas for ethylene dichloride, as well.

Hoffman-LaRoche Corporation stated that "...one of the most significant problems associated with the use of chloroform is its high vapor pressure which makes it extremely difficult to contain during processing. Although compliance with the ACGIH TLV of 10 ppm would be difficult to achieve, the proposed NIOSH REL of 2 ppm would, in our opinion, be technologically infeasible" [Ex. 3-749].

Chloroform is manufactured in a gas phase reaction at temperatures ranging from 350 to 750 degrees C. The major use of chloroform, production of chlorofluoromethanes, also involves reactions carried out at elevated temperatures and pressures. These reactions must be performed in a completely closed system, so routine exposures should be minimal. Given the nature of the production process and the primary use of chloroform, together with the absence of comments from other manufacturers regarding exposures during processing, OSHA concludes that the 2 ppm PEL is technologically feasible.

OSHA recognizes that brief exposure levels of over 2 ppm can be encountered during loading and unloading operations of carbon tetrachloride and chloroform. However, OSHA concludes that the 8-hour TWA PELs of 2 ppm for chloroform and carbon tetrachloride are feasible. Reduction of emissions from tank car and tank wagon openings can be achieved with the use of engineering controls such as vapor recovery systems. Workers should be exposed only for extremely brief periods when attaching or disconnecting lines during loading or unloading operations. Training in and use of proper work practices, in conjunction with proper maintenance or replacement of valves and couplings can reduce both levels and duration of exposures.

The laboratory analysis of samples should be performed under a hood. Overexposures during the collection of samples can be prevented by the installation of sampling boxes if adjustments in work practices are insufficient. Because OSHA allows the use of respirators to prevent overexposures during maintenance activities, feasibility of engineering controls is not a problem for these intermittent activities.

The Vinyl Institute [Ex. 3-624] asserted that modifications would be required to the tank farm vent controls in "a typical EDC/VCM [ethylene dichloride/vinyl chloride monomer] plant...to comply with the proposed regulation for EDC." OSHA concludes that significant exposures will not occur under ordinary conditions in a tank farm because workers are not normally stationed there. The Vinyl Institute also asserted that increased down time of plants would be necessary to clean process equipment of EDC before maintenance work could be performed on that equipment. Because OSHA allows the use of personal protective equipment for maintenance activities, no additional down time or problem of feasibility from this standpoint should be encountered.

The Dow Chemical Company [Ex. 3-741] and the Chemical Manufacturers Association [Ex. 3-874] asserted that the 1 ppm TWA for ethylene dichloride may not be feasible in maintenance, sampling, and loading. OSHA believes that there are engineering controls which can control exposures at these specific operations. However, if engineering and work practice controls cannot reduce exposures to the new PEL, respirators would be allowed.

The feasibility of controlling exposures to carbon disulfide in rayon manufacturing was questioned by the North American Rayon Corporation [Ex. 3-415] and the Inter-Industry Committee on Carbon Disulfide [Ex. 174]. The overexposures are said to occur only when "the windows and hoods are raised to allow access to the machine." The three tasks for which opening the hoods are necessary are for changing spinerettes, for removing filament bundles and to meet product line changes [BASF Corporation, Fibers Division, Exs. 3-674, 125]. BASF claims that in these areas, ambient workroom air cannot be reduced to 1 ppm as a TWA or 10 ppm as a STEL [Ex. 125]. Rayburn H. Dean, BASF Group Vice President, stated that at other times and in all other areas of the plant, exposures are below 1 ppm, although he refused to provide monitoring data on the grounds that it is proprietary [Tr. 8/2/88, pp. 157-159]. Mr. Dean also explained that, referring to the cutting area, "We have some TV monitoring there so that fewer people are in that area. When they are, we have installed this special air conditioned room that you made reference to earlier, down in the spinning room" [Tr. 8/2/88, p. 151]. Manufacturers already have "NIOSH approved respirators that must be used" any time there is a short-term excursion above 20 ppm. [Tr. 8/2/88, p. 136]. "The respirator is the only control available in these three routine operations, to prevent consistent and repeated exposure of the workers to carbon disulfide" [Ex. 125]. OSHA realizes the complexity of this process situation and concludes that the use of respirators during the three aforementioned situations will permit the highest level of protection to workers.

Dow Chemical [Ex. 3-741] and the Halogenated Solvents Industry Alliance [Ex. 3-873] questioned the feasibility of the proposed trichloroethylene (TCE) PEL of 25 ppm in degreasing operations. OSHA concludes that exposures in degreasing can be controlled at the final level of 50 ppm and that exposure data support this position. Both commenters provided information which indicated that, in 1974, 37 percent of open-top degreasers using TCE could maintain 25 ppm and a European estimate from 1980 stated that 50 percent of open-top degreasers and 60-65 percent of closed degreasers could meet a 25 ppm standard. OSHA believes that a considerable amount of overexposure in degreasing is due to inadequate engineering controls or insufficient attention paid to the problem of solvent "carryout." The addition of controls such as chillers, lip exhaust, drying tunnels and covers will reduce personal and ambient exposure levels.

Control of chlorine exposures to the proposed 0.5 ppm ceiling was expected by Dow Chemical Company [Ex. 3-741] to require increased use of respirators by employees engaged in some tasks. Referring to the current PEL of a 1 ppm ceiling, Dow stated: "We have been able to achieve this degree of control in our C12 plants and in the majority of our normal work situations so that respiratory protection is needed only in a limited number of short-term situations." Only one area where these situations occur was mentioned: magnesium production. A NIOSH Health Hazard Evaluation [No. 75-113-249] found a median exposure level of 0.16 ppm for 19 operator's breathing zone samples in a magnesium extraction and chlorination operation. A second NIOSH report [No. 79-40-1381] found a median of 0.2 ppm for 54 samples taken at a chlorine production facility. Only 17 percent of these samples exceeded 0.5 ppm. OSHA concludes that these studies, and the fact that technological infeasibility was not claimed for any specific operations, indicate technological feasibility of a 0.5 ppm TWA PEL and 1.0 ppm STEL for chlorine and that few, if any, additional measures will be necessary to meet these limits.

The manufacturers of cellulose acetate, Tennessee Eastman and Hoechst-Celanese, asserted that the proposed PEL of 250 ppm for acetone is not technologically feasible by means of engineering controls [Exs. 128, 149, 8-54]. Four employee categories were specifically identified in testimony as situations where compliance by engineering controls would be infeasible: filtration workers, dope operators, spinning machine operators and doffers, and parts washing.

While the manufacturers stressed the importance of using acetone as the process solvent, substitute solvents, such as ethyl acetate, could be found for parts washing. In activities such as parts washing, the solvent cannot affect fiber quality, but need only dissolve the cellulose acetate.

Filtration workers at a Kingsport, Tennessee plant dress four or five presses each twelve hour work day [Tr. 8/4/88, p. 142]. According to Mr. Joseph Morton of Tennessee Eastman, "Each press dressing requires about 45 minutes to remove dirty filter media and 45 minutes to apply clean filter media." Assuming that exposures are significant only during the removal of dirty filter media, this would amount to three to four hours of exposure per 12-hour shift. Mr. Morton also observes that "exposure levels of filtration workers frequently are in the range of 500 to 700 parts per million" [Tr. 8/4/88, p. 142]. Exposure monitoring data submitted by the Chemical Manufacturers Association Ketones Program Panel showed filtration workers exposed across a wide range of levels: Four of the 25 samples were in the range 500-750 ppm, 12 of the 25 were 250-500 ppm, and the remaining 9 were below 250 ppm. These data clearly suggest that exposure of filtration workers can be controlled at levels below 750 ppm. Because these samples were taken for the same job title at the same facility, they suggest that the wide range of exposures are due to work practices rather than differences in controls or tasks.

According to the testimony of Mr. Morton, the dope operators are exposed to acetone for about four hours per shift. The primary responsibility of dope operators is to wash the filter cloths in acetone. This exposure generally results in 8-hour TWA levels of 250 to 500 ppm [Tr. 8/4/88, p. 145]. The possibility of using ethyl acetate in place of acetone should be considered for this function, also. Monitoring results provided by the CMA show that most dope operators at the facility where the samples were taken are exposed at levels below 250 ppm. Eleven of the 17 samples were less than 250 ppm and 4 were between 250 and 500 ppm. Again, these samples were taken for the same job at the same facility, suggesting that differences in work practices provide a primary reason for the different levels of exposures. OSHA concludes that improvement to the existing engineering controls and careful attention to work practices would be sufficient to protect the dope operators from overexposure to acetone.

Spinning machine workers can be looked at in three groups: Doffers, acetate yarn spinning machine operators, and tow spinning machine operators. Five 8-hour TWA exposure measurements for doffers were provided by the CMA. Four were in the range 250-500 ppm and one was over 500 ppm. Only one of the 20 samples for yarn spinning machine operators was under 250 ppm. The remaining 19 were between 250 and 750 ppm. Exposure measurements for tow spinning machine operators were split evenly below 250 ppm (26 of 50 samples) and in excess of 250 ppm. Both Tennessee Eastman and Hoechst-Celanese stated that additional engineering controls are not feasible to further protect these operators. Additional local ventilation would cause the fibers to become entangled and complete enclosure would prevent necessary access to the equipment, as well as allowing the possibility of unsafe levels of acetone to build up in the enclosed areas. "...The proposed PEL of 250 ppm [for acetone] is neither technologically nor economically feasible" according to the Chemical Manufacturers Association Ketones Program Panel (the "Panel") [Ex. 98-15].

Based on the evidence submitted OSHA concludes that a PEL of 750 ppm for acetone is not only technically feasible, but is currently being met.

The technological feasibility of the proposed 0.1 mg/m(3) ceiling limit for exposures to nitroglycerin (NG) and ethylene glycol dinitrate (EGDN) is disputed by the Institute of Makers of Explosives (IME) [Ex. 3-749 and Ex. 190]. In this document, the Institute stated: "Reducing workplace levels of NG and EGDN to the proposed ceiling...through the application of administrative controls, engineering controls and/or personal protective equipment is not feasible" (IME's emphasis).

A number of arguments are presented to support this position. First, "Administrative controls (limiting the duration of a worker's exposure) are applicable for reducing...time-weighted averages, but not to the exposure levels based on short-term exposure limits or ceiling limits." Engineering controls are considered infeasible primarily because of safety concerns, such as the collection of explosive materials in local exhaust ventilation ductwork and the dangers of enclosing equipment. IME stated that general dilution ventilation has been effective in meeting the current 0.2 ppm ceiling limits for NG and EGDN. An attempt has been made to estimate the cost and feasibility of engineering controls, but it was concluded that "...the system had a less than 50% probability of successfully attaining a level of 0.01 ppm (0.1 mg/m(3))." On the feasibility of using personal protective equipment to comply with the proposed standards, IME contended that "Air purifying respirators are not generally suitable for use in NG/EGDN-containing atmospheres, and at least one manufacturer, Mine Safety Appliances, specifically warns against their use in such atmospheres." [Ex. 3-749]. NIOSH does not approve the use of canister or cartridge respirators for NG/EGDN because the odor threshold is above the PEL. This means that a worker could be overexposed while wearing a respirator and not be aware of it.

A self-contained breathing apparatus is not considered usable for long-term use because of its weight. Thus, air-line respirators are the single remaining alternative means of achieving compliance. However, the IME contended: "Air-line respirators are not feasible because the air lines restrict employee movement, thereby compromising several areas of operations safety as well as the ease of evacuation in the event of emergencies. In addition, lines trailing behind workers would hinder compliance with the long- standing industry standards for reducing to a minimum level all foreign items which might be accidentally introduced to the production equipment and product." [Ex. 3-749] No studies could be found concerning safety aspects of air line respirators. The lack of studies, complaints, or incidents involving safety problems with air lines despite very common and widespread use, leads to a conclusion that there are no significant problems.

The IME concluded that "...airborne concentrations of NG/EGDN have already been reduced to the practical minimum. Industry hygienists have concluded that reducing airborne concentrations would not decrease worker exposure and any further reductions must be accomplished through the implementation of improved personal hygiene and other workplace practices. The...industry cannot undergo further reductions without dramatically altering the manufacturing process...."

OSHA recognizes the unique difficulties which arise from attempts to control exposures in the explosives industry, but does believe that the final limits can be met through a combination of equipment improvements and respiratory protection. The Institute of Makers of Explosives leaves open the possibility that exposures might be further reduced by process and/or equipment improvements. If compliance cannot be achieved via engineering controls or process improvements, then air-supplied respirators could be employed. Quick-release couplings on the air lines would eliminate problems relating to ease of evacuation in emergencies.

American Cyanamid Company suggested that the proposed standard of 0.03 mg/m(3) for acrylamide is not technologically feasible. To support this position, Cyanamid reports that "NIOSH surveyed the acrylamide monomer manufacturing facilities recently and found that exposure levels were above the 0.03 mg/m(3) level in all facilities" [Ex. 3-961 and Tr. 8/11/88. p. 57]. While exposures above the proposed level may have been found at all of Cyanamid's facilities, all of the personal exposure readings at one of the four facilities surveyed were less than 0.012 mg/m(3). All of the area samples at two of the four facilities were less than 0.015 mg/m(3). The conclusions of the NIOSH Hazard Study are that exposure levels were most dependent on the facility or location where the employee works rather than his job duties and that the primary difference in exposure levels between facilities was due to the background acrylamide air level (see Applied Industrial Hygiene, Vol. 1, No. 3, September 1986, "Evaluation of Occupational Acrylamide Exposures," Bruce Hills and A. L. Griefe, ACGIH, Cincinnati, pp. 148-152). The plant at which the highest exposure levels were measured, the only facility which manufactured dry acrylamide, has since closed. The range of personal exposure measurements at the remaining three facilities was 0.001-0.132 milligrams per cubic meter. Based on the data and conclusions of the NIOSH field studies described in the Hazard Evaluation, OSHA concludes that the PEL for acrylamide of 0.03 mg/m(3) is technologically feasible.

U.S. Borax submitted information regarding expenditures they have made on environmental controls to reduce exposures to borates. Since 1970, they have spent $7.5 million at the Boron, California, facility. Although some of these expenses have been related to the mining operation under the jurisdiction of the Mine Safety and Health Administration (MSHA), the remainder are related to operations which are of concern to OSHA. The range of dust levels in the fusing department has been reduced to 0.63-50.54 mg/m(3) and in the shipping department to 0.25-15.56 mg/m(3) [Ex. 3-744]. It is not clear whether further reductions could be achieved, and U.S. Borax does not address this issue. Part of U.S. Borax' efforts to control dust related to paving some areas to reduce background dust levels. This effort has apparently been at least partially successful, as evidenced by the minimum values in the range of exposures presented above. It appears that the 10 mg/m(3) PEL level for borates should might be achievable under most circumstances. Further reduction in borate dust levels might be achieved through the installation of additional engineering controls such as dust collection systems for bagging and packaging, additional dust collection systems at critical release points and further reduction of background dust levels. If additional reductions do not achieve the required levels, the use of respirators will be necessary to protect workers.

SIC 29 - Petroleum Refining

In order to assure the quality of petroleum products and determine quality of waste streams, petroleum refiners must sample their process streams periodically. As with maintenance, workers that sample process streams are at risk of being in close contact with a variety of chemicals. Controls for this operation involve sampling boxes that vent gases and vapors away from the operator and/or shield the operator from accidentally splashed or spilled material.

Process stream samples are taken to the laboratory to determine if their qualities lie within acceptable limits. As laboratory workers perform analyses, they can be exposed to various organic and inorganic chemicals if appropriate engineering controls are not in place or if proper procedures are not used. Exposure controls include exhaust fans and laboratory ventilation hoods. In general, this industry has extensive control technology in place for the primary processing equipment. Closed processes with few exposed workers are predominant due to the requirements of process operation at elevated temperatures and pressures.

SIC 30 - Miscellaneous Plastic Products

The Styrene Information and Research Council (SIRC) identified open-molding processes (i.e.,processes in which styrene, frequently in combination with fibrous glass, is sprayed or rolled into a mold manually) as the type of process most likely to have difficulty meeting the proposed PEL for styrene [Ex. 3-742, p. 3 and Tr. 8/3/88, p. 5-94]. In SIC 3079, open-molding processes were identified as being used in the production of underground storage tanks, lavatory castings, tubs and spas, and cultured marble products [Ex. 3-742, p. 105 and Tr. 8/3/88, p. 5-181]. Other products that are made using open-molding processes include bridges for military vessels. [Tr. 8/3/88, p. 5-188], planters [Ex. 3-742, Attachment 2, p. 15], benches [Tr. 8/3/88, p. 5-195], and chimney stacks [Tr. 8/3/88, p. 5-188].

Worker exposures to styrene occur principally in two process steps in the open-molding process: gel-coating and lamination [Ex. 3-742, Attachment 2, pp. 17-19 and Tr. 8/3/88, pp. 5-131 to 5-133. Gel-coat is a pigmented resin of polyester resin-based paint. The application of gel-coat is similar to the application of paint and is normally done using an air atomizer or airless spraygun. Lamination may be applied using either hand layup or hand sprayup. In hand layup, workers place a layer of fiberglass matting directly onto the mold and secure the fiberglass with a layer of resin, which is normally applied with rollers or brushes. In the sprayup process, a chopper gun chops fiberglass roving into pieces and sprays resin at the same time, so the two converge and are sprayed onto the mold simultaneously.

The most extensive data source on exposures to styrene in this industry sector is a study conducted by the State of California's Division of Occupational Safety and Health (DOSH) [Ex. 3-742, Attachment 2]. This study reported the results of an in-depth industrial hygiene survey of styrene and other hazardous workplace exposures in the fiberglass/reinforced plastics industry. A total of 141 workplaces were inspected, and 379 of the 2600 workers employed in these workplaces were sampled over a full workshift [Ex. 3-742, Attachment 2]. The report also recommends the best control measures to minimize hazardous exposures; the focus of the study was on large open-mold sprayup/layup operations, because earlier research had shown that these open-molding operations had the highest exposures of all operations in these workplaces [Ex. 3-742, Attachment 2]. Styrene exposures (8-hour TWAs) at these processes ranged from 0.2 to 288 ppm; the 8-hour TWA arithmetic mean and the median for these sample results were 43.0 ppm and 34 ppm, respectively [Ex. 3-742, Attachment 2].

In a comparison of worker exposure levels by industry, the California OSHA study showed that the geometric mean exposure levels were highest in tub/ shower manufacturing facilities (53.6 ppm), followed by camper manufacturing facilities (41.0 ppm), spa manufacturing facilities (25.8 ppm), miscellaneous manufacturing facilities (22.0 ppm), and tank manufacturing facilities (12.7 ppm). Operations ranked according to percentage of styrene exposures above 100 ppm as an 8-hour TWA (the former OSHA limit) were: tub/shower manufacturing (19 percent); spa manufacturing (11 percent); camper manufacturing (6 percent); miscellaneous plastics manufacturing (4 percent); and boat and tank manufacturing (none).

The industrial hygienists who conducted this study initially believed that working on large molds, such as those involved in making boats (see discussion for SIC 37) or tanks, would result in the highest styrene exposure levels, because the mold almost surrounds the worker, making a kind of confined space. Workers engaged in boat and tank manufacturing, however, had the lowest overall exposure levels, while the tub/shower and spa manufacturing sectors had more workers exposed above 100 ppm. A partial explanation for these differences in styrene exposure levels in various industry sectors is caused by differences in work production rates according to California OSHA. In boat manufacturing, for example, sprayup operations are performed at a slow and intermittent rate while tub/shower manufacturing is conducted at an assembly-line pace [Ex. 3-742, Attachment 2]. Industry representatives also believe that production volume plays a large role in determining styrene exposure levels. Jack Winnick, general manager of Gold Shield Fiberglass in Fontana, California, testified that plants in Western Europe can achieve much lower PELs than can plants in the United States because "[t]he volume of resin throughput and products produced is a mere fraction of the throughput in U.S. facilities..." [Tr.8/3/88, p. 5-114].

California OSHA found, however, that the factor determining whether or not 50 ppm TWA is currently being reached in facilities producing reinforced plastics products is the degree to which effective controls have been implemented. The California OSHA researchers and Diane Factor of the AFL-CIO both reported that, in all cases where companies had implemented effective control measures, employee styrene exposure levels were below 50 ppm [Ex. 3-742, Attachment 2, and Tr. 8/4/88, p. 6-64].

There is some question regarding the representativeness of the California studies of conditions found elsewhere in the nation. First, SIRC notes that winter climates in the northern-tier states may present additional problems in achieving the proposed PELs [Ex. 181A, p. 38]. Furthermore, the findings of the California study were qualified by its authors as follows:

This study was conducted in a CAL/OSHA compliance mode: This represents two problems: (1) industrial hygienists do not have the luxury of making frequent visits to any one site, and (2) employees have an understandable desire to minimize actual exposures by various means...in order to avoid CAL/OSHA citations and fines [Ex. 3-742, Attachment 2, p. 31].

OSHA appreciates the critiques of the CAL/OSHA study. OSHA found that employers would need to employ a flexible compliance strategy during manual layup/sprayup operations to achieve the proposed limits in boat-building facilities (see Technological Feasibility discussion for SIC 37). The California study indicates that employee exposures to styrene during manual layup/sprayup operations in facilities in SIC 30 are even higher than those for boat-building facilities. Thus, there is uncertainty about the technical feasibility of achieving the 50-ppm TWA and 100-ppm STEL limits exclusively by implementing engineering and work practice controls during manual layup/sprayup operations in SIC 30. Respirators as well as engineering and work practice controls may be necessary to achieve these limits in some operations. OSHA concludes, however, that for most operations in SIC 30 where styrene is used, the revised TWA and STEL limits are technologically feasible.

Daniel Boyd, representing the SIRC, also commented that the mixtures formula described in 1910.1000(d)(2)(i) would necessitate reducing employee exposures to well below 50 ppm. He stated that:

Since the reinforced plastics environment consists of a number of chemical constituents, the allowed exposure to these various chemicals must be calculated through the mixture formula....OSHA's application of the mixture formula will require limits well below 50 ppm for styrene and [the proposed PEL of] 250 ppm for acetone [Tr. 8/3/88, p. 5-97].

The SIRC thus argued that, with the mixtures formula, "OSHA has proposed a rule [for styrene] that would impose an exposure level lower than it has accounted for in its feasibility analysis" [Ex. 181A, p. 44].

OSHA does not agree with the SIRC that the mixtures formula requirement will have a substantial impact on the ability of employers to comply with the 50 ppm PEL for styrene. Traditionally, OSHA does not apply the mixtures formula in most cases where multiple exposures occur without careful consideration of the additive or synergistic effects of the specific substances involved. According to OSHA's Field Operations Manual (FOM):

The use of..[the mixtures formula] requires that the exposures have an additive effect on the same body organ or system. Caution must be used in applying the additive formula, and prior consultation with the Regional Administrator is required (OSHA FOM, Chapter IV, Section 6(e)(i)).

Thus, in the case of styrene and acetone, which are both used in the reinforced plastics industry, OSHA does not believe that the mixture formula rule specified in paragraph (d)(2)(i) of the final rule would necessarily apply, because styrene is principally a narcotic agent that acts on the central nervous system and acetone is primarily an irritant that acts on the eyes and respiratory passages at concentrations at or below the final PEL. These substances therefore cannot be considered as having an additive effect. Consequently, OSHA has based its feasibility assessment for styrene and acetone in the reinforced plastics industry on the availability of the engineering and work practice controls necessary to achieve the PELs for these substances individually.

Carbon disulfide is a solvent used in the production of cellulosic food casings. It is reacted with cellulose to make xanthate and is slowly released during subsequent steps of production. The process currently used is the only known process for producing cellulosic food casing, and carbon disulfide is the only known solvent for this process [Ex. 8-45].

The feasibility of controlling exposures to carbon disulfide in the manufacturing of cellulosic food casings was questioned by representatives of the producers of these products, Viskase Corporation and Teepak, Incorporated [Exs. 33, 162, 3-753, 8-19, 8-45, Tr. 8/2/88, pp. 4-201 -217]. These commenters noted that, in three specific operations, it is necessary to open the machinery to perform manual operations (unloading the baratte, aligning strands in the cabinet, and manual puncturing of the casings). When unloading the baratte, manual raking is required because of the light and sticky characteristics of xanthate, the parent compound. Operator access is required to keep the strands of product properly aligned within the extrusion apparatus (or "cabinet"). Manual puncturing of the casings is required downstream of the extrusion nozzles [Ex. 8-45]. Personnel performing all three of these operations must open the process machinery while performing these tasks; currently, personnel wear air-supplied hoods to protect against the carbon disulfide excursions above 20 ppm associated with thee operations [Tr. 8/2/88, p. 4-228]. Operator access is essential to assure casing quality [Ex. 162]. An engineering study conducted in one of Viskase's plants concluded that "it is highly unlikely that the 1 ppm [level] could be obtained at [these] three...routine operator tasks," even if "the most extreme measure that can be visualized as an effort to reduce the concentration" was employed [Ex. 8-45]. Teepak stated that "no one...has developed a system or knows of any engineering controls that...[are] capable of reducing CS(2) levels in the casing industry to the 1 ppm level proposed by OSHA," adding that Teepak had recently redesigned and rebuilt much of its plant using the best available technology [Ex. 162]. Commenters repeatedly stressed [Exs. 33, 162, Tr. 8/2/88, pp. 4-201 to 4-217] that feasibility was a problem only for these manual operations; thus, OSHA concludes that the use of respirators as well as engineering controls and work practices, may be necessary for unloading xanthate from the baratte, aligning strands in the extrusion cabinet, and manual puncturing of casings at the extrusion nozzles unless OSHA can demonstrate that engineering controls and work practices alone can achieve the PEL limit for carbon disulfide.

The Polyurethane Manufacturers Association (PMA) stated that the 0.02 ppm PEL for 4,4'-methylene bis (2-chloroaniline) (MBOCA) is technologically feasible and is already being achieved in many facilities [Ex. 3-683]. MBOCA is used as a fixative in producing castable polyurethane. The chemical is no longer produced in this country, but is still widely used to produce castable polyurethane products. PMA [Ex. 3-683] also stated that no substitute for MBOCA has been found that matches its physical properties and processing characteristics at a competitive price.

According to PMA, worker exposures to MBOCA occur chiefly during transfer operations. PMA stated that "once the melted MBOCA is mixed with prepolymer, there is no risk of employee exposure to MBOCA" [Ex. 3-683]. The industry has developed a number of methods to control employee exposure to MBOCA during transfer operations, including the use of isolated rooms, laboratory hoods or glove boxes, and vacuum transfer systems that carry MBOCA from drums to the melters in closed, automated systems. PMA also stated that "based upon considerable workplace monitoring [conducted] since the 1970s, it is apparent that an employer who observes the recognized industry practices for the use of MBOCA and who monitors the results...will feasibly comply with the proposed TWA level [of 0.02 ppm]" [Ex. 3-683]. Therefore, OSHA concludes, based on the proven effectiveness of currently available technology that a PEL of 0.02 ppm for MBOCA is technologically feasible.

SIC 31 - Leather and Leather Products

During a site visit performed by OSHA to a shoe production facility [Ex. 11 -Firm 7], an overexposure to 2-butanone at the outer sole cementing operation was found. A small exhaust system used at the operation had inadequate air movement to reduce exposure. The exposure exceeded the current PEL of 200 ppm, as well as the final rule STEL of 300 ppm. The length of the exposure was 3.5 hours. The total cost for local exhaust ventilation to the five affected work stations, as estimated by Clayton Environmental Consultants, would be $20,000. This cost figure is based upon a flanged 4 foot by 12 foot exhaust hood with a capture velocity of 100 feet per minute (fpm). The flow rate is estimated to be 775 cubic feet per minute (cfm) per work station. By employing these control measures, worker exposures to 2-butanone, as well as to solvents in general, will be reduced.

The site visit firms did not have toluene overexposure. In general, toluene exposures can be decreased by revising standard work practices to reduce the contact time between leather and toluene.

SIC 32 - Stone, Clay and Glass Products

In batch mixing of raw materials for glass production (SICs 321 and 322), OSHA found that drysweeping and/or the use of compressed air for cleaning may contribute substantially to the employees overall exposures. By substituting vacuum cleaning systems, worker exposure can be reduced. The Brick Institute of America (BIA) stated that limited success has been achieved by member companies in controlling clay and shale dust exposures. Despite the companies' efforts, employees are still required to wear respirators in some areas of the plants. Although local exhaust ventilation has been installed at work stations, the nature of the job requires workers to leave the area of their work station. Furthermore, the moisture content of the raw materials "creates substantial maintenance problems for the control equipment." The BIA concludes "...it simply is not possible to reduce dust levels any further using known technology" [Ex. 130].

OSHA conducted site visits to manufacturers of both cement blocks [Ex. 11 - Firms 2, 4] and a manufacturer of unglazed floor tiles [Ex. 11 - Firm 12]. These firms have processes analogous to brick manufacture. No overexposures were found. In addition to good housekeeping measures, one of the principal means of controlling dust exposure was the use of wet materials. Dry material hoppers were located outside the building or in locations above the work floor. Mixers were generally fully automated. Workers who were required to work inside the mixer (after proper lockout procedures) using pneumatic hammers to remove hardened materials, used local exhaust tubes, and wore respirators. OSHA concludes it is technologically feasible to control dust exposure in this industry.

SIC 33 - Primary Metals Industries

The American Iron and Steel Institute [Exs. 3-1123, 72, 129, 188] stated that the proposed PELs were not technologically feasible if retrofitting were required, and that most operations produce intermittent exposures where respiratory protection equipment is more appropriate. The AISI provided no support for their statement that retrofitting is not technologically feasible, merely stating that some controls can only be installed when a plant is built or modernized. However, they provided several examples of the cost of such retrofits, stating that the costs rendered the retrofits infeasible. This, however, does not support a finding that the technology does not exist to control the exposures.

During site visits, OSHA observed a wide variety of controls in place in this industry. How effective these controls are cannot be known for certain. When AISI agreed to assist OSHA in arranging site visits in this industry, they stipulated that the OSHA contractors could collect no air samples during the site visits. AISI submitted exposure data for different operations, but provided only the lowest and highest values for each chemical in each operation [Ex. 129]. When ranges are used, it is not possible to discern where most samples fall to assess potential feasibility. Where the highest value for a chemical is at or below the new PEL, the new PEL is clearly feasible. This is true in a number of cases, and demonstrates that controls are available which will maintain exposures below the new PEL for those chemicals. The exposure data does indicate that the STEL for sulfur dioxide cannot be regularly achieved with engineering and work practice controls in blast furnace operations and at sulfur plants. In addition, there is some evidence that the ceiling limit for carbon monoxide cannot be regularly achieved with engineering and work practice controls at blast furnace operations, vessel blowing, basic oxygen furnaces, and sinter plants. There is no evidence to the contrary in the record for with of these two substances. OSHA, therefore, will permit more flexibility in the use of respirators for these operations. The burden of proof will not be on the employer to demonstrate that compliance with engineering and work practice controls are infeasible in a compliance action for exposure to the STEL for sulfur dioxide and the ceiling for carbon monoxide at these operations.

AISI also provided a list of occupations and the related duties where exposures are intermittent. Many of these would be considered cleaning and maintenance [Ex. 72]. Where exposures are brief and intermittent, or where they are related to cleaning and maintenance, respiratory protection may be the appropriate control technology in accord with OSHA's traditional policies.

SIC 331 - Basic Steel Products

OSHA, through its contractors, has conducted site visits of various operations associated with steel manufacturing. During these visits, OSHA observed that engineering controls were in use. Due to a pre-visit stipulation of the American Iron and Steel Institute, OSHA was not able to monitor exposures at any of these operations.

In a site visit to a sintering plant, OSHA observed the application of a hood and local exhaust at the end of the sintering conveyer, a transfer point for sintered material. Also, at the same plant, local exhaust piping on the pug mills and the sinter air cooler was in place. The emissions from these sources were directed to a centralized baghouse [Ex. 129 - Firm 28].

At hot strip production facilities [Ex. 129 - Firms 29, 31], workers controlling the rolling process were positioned in air conditioned control stations or pulpits. Workers engaged in the coiling and marking area were provided with dilution ventilation.

At Basic Oxygen Furnace (BOF) facilities [Ex. 129 - Firms 37, 38] emissions generated during desulfurization, deslagging and oxygen injection were vented to an electrostatic precipitator and/or baghouse. BOFs also emit carbon monoxide as a byproduct. Some BOF processes use this as a means of controlling the metallurgical reaction. This is controlled through the exhaust system. Continuous CO monitors are used to alert workers to peak or emergency conditions.

During a visit to an electric arc furnace operation [Ex. 129 - Firm 39], emission control equipment was an integral part of the furnace's rotating roof. Contaminant generation during this operation was vented to an electrostatic precipitator. The building in which this process was conducted had been modified to incorporate roof level hoods and ducts that carried escaping contaminants to a centralized baghouse.

At a coke oven gas processing facility [Ex. 129 - Firm 32], operations were carried out in enclosed vessels or process equipment (similar to those found in chemical processing facilities) that provide protection from continuous direct exposure. OSHA believes that compliance with the proposed PELs can be achieved since exposures are primarily the result of fugitive emissions and operational upsets.

At a blast furnace operation [Ex. 129 - Firm 27] which was visited, the firm indicated its concern over the proposed PELs for carbon monoxide, sulfur dioxide, and calcium oxide. (Iron oxides were also generated in these operations but it was not clear that exposures to iron oxide were problematic. The final rule retains the existing limit for iron oxide.) Blast furnaces operate under positive pressure and extremely rigorous conditions. These conditions do have a severe effect on the refractory lining of the furnace wall. Over time contaminant release, particularly carbon monoxide, will occur. Thus blast furnaces of older design or furnaces reaching the end of their life cycle will tend to have greater emissions of air contaminants. During the tapping of the furnace, workers are exposed to iron oxides, carbon monoxide, sulfur dioxide, and calcium oxide as hot metal pours into the transfer car via runners on the floor. In one site, these runners were covered to reduce emissions.

Workers typically move between areas of high exposure in the manufacturing area, and areas of low exposure in air-conditioned or heated control rooms or "shanties." Exposures are intermittent. Processes at which they work also have episodic periods of air contaminant emission. This dual variability suggests that respiratory protection may be needed to control worker exposure to such metallurgical air contaminants as carbon monoxide, sulfur dioxide, iron oxide, and calcium oxide. See the discussions on pages 2526 and 2652 for the specific operations and chemicals where respirators may be used unless OSHA can demonstrate engineering controls are feasible.

SIC 332 - Iron and Steel Foundries

OSHA conducted a site visit to a gray iron foundry [Ex. 11 - Firm 13] engaged in the manufacturing of gray and ductile iron castings. Exposure samples were taken in the grinding process. The result of the sampling disclosed an exposure level of 39.0 mg/m(3) TWA for iron oxide which is above both the current and final PEL of 10 mg/m(3) TWA.

Clayton Environmental Consultants recommended a number of actions that can be taken to reduce iron oxide dust levels. A mechanical shakeout and automatic sand handling system complete with dust collection can be implemented to substantially reduce dust levels. The mechanical shakeout would consist of a 16 square foot enclosure in which the molds and castings can be manipulated and then brought back out for further processing. The existing muller hood should be maintained or upgraded to produce 900 CFM of local exhaust ventilation. It was also recommended that dust collection and make-up air systems would be needed to replace exhausted air. For this purpose, make-up air should be ducted for release near the work station and workers should be provided with movable diffusers or a means of controlling airflow at the work station. The purchase of a rider-type sweeper was also recommended as a means to control dust levels.

Clayton also recommended controls to reduce grinding exposures. Local exhaust ventilation can be installed on pedestal grinders. Grinders of this type should be exhausted at 1,000 CFM assuming 16 inch wheel diameters. For hand-held grinders, a 2 foot by 3 foot ventilated table/bench is the recommended control. A 200 CFM/foot(2) of grinding bench area exhaust rate is the estimated requirement for this application.

Additional controls applicable to foundries were found during OSHA's review of this industry. These controls either individually or in combination should generally control exposures to the final PELs.

* The arc air process in steel foundries (SICs 3324, 3325) was used during the processing of steel castings to control fumes. In order to ventilate the arc air booths, fumes were exhausted through the back of the booth and fresh air was supplied from above and behind the operator.

* Steel foundries (SICs 3324, 3325) used an overhead canopy hood during the induction melting of steel to control fumes. The hood consisted of sheet metal barriers extending down from the roof to the top of the hot metal ladle monorail. Thermal drafts carried the fumes upward into the hood where they were exhausted by ventilators. Mancooler fans behind the workers pushed some fumes under the hood.

* Emissions during the oxy-acetylene torch cutoff of risers from steel castings was encountered in iron and steel foundries (SIC 332). Castings were cut in a specially designed booth with a rear exhaust flow and a frontal air supply flow. Air pressure from the cutting nozzle of the torch was directed toward the rear exhaust port for effective dust and fume control.

* Fume control of a sandwich-type inoculation in iron foundries (SICs 3321, 3322) was achieved through the use of a commercially available canopy hood. The fume-laden air was exhausted through mobile duct work and cleaned by a fabric collector before being discharged into the surrounding environment. The hood tilted with the furnace so that it always was directly over the ladle for fume capture.

* Fume, dust, and gas control from the melting of iron (SICs 3321, 3322) in an arc furnace was achieved by the installation of a hood. The exhausts collected by the hood were filtered by cloth filters before being released into the external environment.

* Control of dust and gas emissions from phenolic urethane cold box coremaking in iron foundries (SIC 3321, 3322) included local exhaust ventilation which provided negative pressure at the core box. Parting line gaskets, blow seals, and stripper pin o-rings were regularly maintained for emission control. Exhaust outlets captured excessive dust.

* In an iron foundry (SICs 3321, 3322), hot combustion gases were exhausted and flowed through an after burner, cooled, and then passed through a dust collector. Tapping emissions were captured by a canopy hood. General ventilation was provided by mancooler fans.

The UAW [Ex. 197] listed additional feasible control measures in foundry and other metallurgical operations. These include:

(a) adequate make-up air;

(b) supplied air islands for operator stations (laminar flow, down draft make up air supply units);

(c) tempered (cooled) make up air to reduce the need for high velocity air for heat stress relief;

(d) process arrangement to remove loose sand from castings after shakeout before they are finished;

(e) maintenance of enclosures and exhaust volume for sand handling equipment to prevent emission of dust;

(f) reduction of silica burn-in on castings to reduce exposures at cleaning and finishing operations;

(g) use of wet methods.

SIC 333/334 - Primary and Secondary Non-Ferrous Metals

In its review of technological feasibility in the non-ferrous metals industry, OSHA identified the following examples of engineering control and work practice measures:

* Control of emissions from aluminum ore handling and storage (SIC 3334) was addressed with an unloader which uses movable vacuum nozzles to remove alumina and coke from barges. The ore was moved on an enclosed conveyer which was equipped with air exhaust hoods at loading and transfer points. The operator can be situated in an air conditioned cab.

* Reduction of alumina dust emissions during ship unloading (SIC 3334) was achieved by automating and controlling operations from an enclosed control booth. Furthermore, mixing operations were hooded and exhausted.

* During anode rodding in prebake plants during primary aluminum production (SIC 3334), spent butt remover, butt crushers, cast iron remover, and shot cleaner were exhausted to a bag filter dust collector. Use of induction furnaces and exhaust hooding reduced metal fume exposure during melting. Hoods and slotted hoods were also used. The operator can maintain controls from an enclosed console.

* Control of air emissions during potline operations of aluminum smelters (SIC 3334) was achieved through the use of potroom ventilation and automated processes such as the use of hooding which consists of curved and ribbed shields, the employment of a dual draft system, and an exhaust system which leads to a dry scrubber. Other control methods included hooding with rigid air-operated doors which exhausted the emissions through air takeoffs to an expanding duct exhaust manifold which, in turn, was exhausted by a fan. Furthermore, computer-controlled systems existed which could automatically perform production functions without requiring workers to open pots or hood shields above pots.

* The mercury cell process may be used in aluminum smeltering (SIC 3334) to produce chlorine gas from brine water. To reduce chlorine gas exposure as a result of this production, the diameter of the brine header was increased to accommodate the gas phase above the liquid phase; the number of cells in the system was increased; the ph of the brine was adjusted; the compressor controls were modified to accommodate surges in pressure; inlet box covers were replaced with better covers; and the brine feed nozzle flange was modified.

* Several engineering controls have been recommended for copper smelting locations (SICs 3331, 3341). A preventive maintenance program can be developed and implemented to insure that ventilation and conveyer systems are operating properly. Dead beds can be installed in chutes to break the fall of material and reduce the level of dust generated. Pneumatic aerators can be installed to eliminate the need for manual air lancing in bins and chutes. Industrial vacuum systems can be used.

* Collection hoods can be installed at each conveyer transfer point at copper smelter sites (SICs 3331, 3341) to control copper particulate. Primary copper smelting conveyer skirting can be properly adjusted, and fingers installed at discharge points. Inspection doors should not be left open, and the lunchroom/breakroom should be located outside of the reclaim building. General measures throughout copper smelting plants (SICs 3331, 3341) to control copper dust emissions included: using local exhaust ventilation for localized sources, and general exhaust ventilation for areas with unidentifiable sources; enclosing conveyer belts and transfer points; enclosing the air conveying system for the transfer of flue dusts; enclosing workers' operating vehicles; installing secondary hoods on converters; prohibiting the blowing out of converters while on stack; performing preventive maintenance on balloon flues; not allowing converters to remain rolled out for extended periods of time; and providing cleaning rooms with filtered, tempered, positive pressure air. When hauling slag from the metal smelting operation, slag can be granulated after skimming with high velocity water; a chemical dust suppression system can be used when crushing any cooled slag; and the slag crew can ride downwind from fumes. Further engineering controls include constructing pulpits for operators; close-coupling the ventilation system to the Larry car; using dead beds in calcine loading; enclosing a portion of the building to block wind; and vacuuming the superstructure of the Larry car and any spills.

* Controls used to decrease exposures to arsenic, dust and sulfur dioxide at primary copper and lead smelters (SICs 3331, 3332) included upgrading the present ventilation systems; operating electric furnaces at negative pressure; eliminating air lancing as a method of removing concentrates from receiving hoppers; using pneumatic aerators or belt wipes; using wet techniques in storage; reclaiming concentrates; and improving general housekeeping.

* Exposures to lead, cadmium, and arsenic at lead and copper smelters (SICs 3331, 3332) were reduced by the replacement of old sintering machines with ones equipped with dust and fume controls and by placing a cover over the charge hole when slag was not being charged into the reverberatory furnace.

* Use of a multipurpose crane with an enclosed cab reduced operator exposures to air emissions at carbon bake plants (SIC 3334). The cab was supplied with filtered conditioned air. The crane was equipped with a vacuum system which could aspirate cake from ovens and separate fines.

* Controls for exposure to soluble platinum salts in precious metal refining (SIC 3339) included local exhaust ventilation used in jaw crusher and recovery sampling, maintenance of a closed system in refinery through use of glove box filters, the use of borohydrate solution to wash down spills and reduce salts to insoluble platinum metal, and mandatory showers and daily clothing changes.

* Controls for the primary non-ferrous metals industry (SIC 333) included local exhaust ventilation systems; general dilution ventilation; covers, hoods and exhaust systems for belts, material handling and transfer systems; enclosure and exhaust of sinter machine area; local exhaust and dilution ventilation for the reverberatory and refinery areas.

* The reduction of exposures to inert cadmium and silver dust during a ball mill operation was accomplished by building and equipment process changes such as local exhaust ventilation, hood enclosure of process or worker, and air cleaning equipment.

* In the secondary smelting and refining of non-ferrous metals (SIC 334), particulate emissions from a dross mill were reduced by making modifications to the dust collection system and to air volumes drawn through the baghouse. Engineering controls used include increasing fan efficiency through the use of sheaves and belts, installing water sprays on crusher infeeds, running new pipe to localized dust areas, installing additional cleanout ports, and replacing the top of the baghouse.

* Employee exposure to nuisance dust from zinc smelters (SIC 3333) was controlled by replacing the dross handling operation with a dross mill. The crusher was replaced with a rotating mixer, thus eliminating fugitive dust from this part of the process.

Asarco, Inc., questioned the technological feasibility of achieving the proposed PELs for sulfur dioxide. In its written statement to the docket [Tr. 8/15/88, pp. 120-124], Asarco provided several examples of the nature of engineering controls that have been installed at its plants. "Asarco's copper smelter in Hayden, Arizona, has been modernized with the installation of an Inco flash smelting furnace, as well as the installation of control devices, such as secondary converter hoods. Additional controls, including secondary converter hoods with an air current design have also been installed in Asarco's copper smelter at El Paso, Texas." Asarco maintains "despite these controls, however, SO(2) concentrations for a number of job classifications at Asarco's copper smelters exceed or may exceed the proposed TWA of 2 ppm. Moreover, it appears that most smelter jobs in molten areas would frequently exceed the proposed STEL of 5 ppm, because of the episodic nature of smelting operations. Asarco is not aware of any combination of engineering and work practice controls that can feasibly reduce exposures to the levels required by the standard."

Magma Copper Company, in written testimony [Tr. 8/12/88] has also expressed concern on [Ex. 3-91, pp. 92-105] the achievement of the proposed standards for SO(2). Magma is in the process of installing new emission controls at its smelter at a cost of $132,000,000. The elements of this retrofit include "a new state of the art Outokumpu flash furnace obviating three existing reverberatory furnaces. The retrofit has also an improved converter gas handling system as well as a new secondary gas collection system. The smelter has local ventilation to all areas that have historically been sources of SO(2) emissions, and as such, the smelter should have limited problems with fugitive emissions...." "Additionally, the converters have two separate local ventilation systems. The primary system collects the highest concentration of SO(2) and supplies an acid plant. The secondary local ventilation system collects gases during some phases of the converter roll out operation." Magma has stated that the flash furnace "was placed on line in July of this year" and "start-up is estimated to be completed with normal operations in place by November 1, 1988. Therefore, a comprehensive survey of our new engineering control system is not feasible at present." However, Magma has estimated from past data and "for various configurations" that the 5 ppm STEL "is not likely to be met."

Asarco has also submitted as an attachment to its written statement a portion of its 1977 post-hearing brief that discusses the problems of SO(2) control. In it, Asarco refers to the report "Environmental Conditions in U.S. Copper Smelters" by William L. Wagner of NIOSH as evidence of the need to use respirators. Quoting from the report, Asarco cites "In most smelters the use of respirators is essential on charge floors of reverberating furnaces, in the green feed galleries, tripper decks above these furnaces or any areas above these furnaces. In these areas, the SO(2) concentrations varied from non-detectable levels to many hundreds of parts per million...." In addition, Asarco also relies on the testimony of Mr. Wagner at the 1977 hearings in which he addressed the then-proposed "ceiling limit" of 10 ppm, "for most parts of the smelter." "There are a number of areas where you could get concentrations of sulfur dioxide for periods of time greater than 10 parts per million." According to Mr. Wagner, concentrations exceeding 10 ppm could easily last 15 to 20 min. as this discussion indicates, there are many engineering and work practice controls available to reduce exposure to SO(2) and the other contaminants present. They will frequently be able to control exposure to 2 ppm. However, for some operations, feasible engineering controls may not be available. OSHA will accept the use of respirators for these operations in conjunction with engineering controls unless OSHA can demonstrate that engineering controls and work practices alone can achieve the 2 ppm PEL.

Brief peak exposures will occur over 5 ppm in several areas of a lead or copper smelter. Good work practices will curtail many of these. However, respirators may be appropriate in smelters along with other controls to control peak exposures unless OSHA can demonstrate that engineering controls and work practices alone can achieve a 5 ppm STEL.

SIC 336 - Non-Ferrous Foundries

Fumes were controlled during the casting of bronze in foundries (SIC 3362) through the use of enclosing hoods. A mobile hood exhausted the ladle at all hot metal transfer points. Flexible ducting connected the hood to a traveling exhaust carriage.

SIC 339 - Miscellaneous Primary Metals Products

Manufacturers improved dust control using closed screw conveyers in the transport and manufacture of iron powder (SIC 3399). Open conveyer belts were changed to a closed screw conveyer system. Duct work was totally replaced. Local exhaust was provided for the rotary screens. New baghouses and electrostatic precipitators were also installed. OSHA visited a manufacturer of metal alloy powders [Ex. 145, Attachment A]. Although overexposures to the current PELs were not found, a reduction to the new PELs would result in overexposures. To reduce cobalt dust exposures below the 0.05 mg/m(3) level, additional monitoring should be conducted to verify the need for engineering controls. The following measures were recommended and determined to be necessary: 1) use of an exhausted booth for developmental screening, 2) routing air discharged from the dust collector associated with the vortex classifer to the outdoors or into the plant's main dust collection system, 3) providing exhaust ducts to be connected to atomizer drums during cleanout periods, and 4) discontinue the practice of dry-sweeping the floors and either acquire a vacuum sweeper truck or use the central vacuum system more extensively.

SIC 34 - Fabricated Metal Products

Control of copper dust at a cookware manufacturing plant (SIC 3469) was addressed by unclogging the ventilation system, repositioning cooling fans, and instituting weekly ventilation system inspection and maintenance programs.

A plating shop (SIC 3471) uses extensive local exhaust ventilation to control worker exposure. Each part to be plated undergoes some surface pretreatment. This can consist of shot-peening, abrasive blasting, degreasing, wax or tape masking and other treatments. Parts are manually placed into the tank using an overhead hoist for large parts. The tanks are set on top of concrete ducts. The floors of the shop and the aisles between the tanks are reinforced concrete, however the area around the perimeter of the tank is open to the basement and covered by steel grating. The ducts are connected to a fan on the roof of the building.

The largest of the hard chrome tanks, holding over 1000 gallons of plating solution, has a two sided lateral exhaust ventilation system. The slot on each side consists of a series of seven slots. The slots are set back from the edge of the tank but an overhanging hood extends to the edge of the tank. A second tank has both a two sided slot ventilation system and a cover. This two piece cover is hinged to a ventilation manifold and extends beyond the front and rear edges of the tank.

Arc welding is performed in many SICs as an auxiliary process and in several industries such as fabricated structural metals (SIC 3441), as the principal process requiring engineering control. During the welding process, temperatures are sufficiently high to vaporize some of the base material of the electrode and produce large quantities of fumes containing the elements in the electrode and the base metal. Thus welders and other workers in the vicinity are exposed to mixtures of fume-sized particulates and both irritant and toxic gases which in combination may have additive or synergistic physiological effects.

Differences in worker exposure are attributable to a variety of factors including type of welding helmet worn, position of the welding operator, the work environment, arc time, and the availability and performance of ventilation equipment. Arc time varies greatly due to differences in work schedules, set up times, and the sizes, shapes and types of tasks. Tasks can vary from short-term repairs conducted irregularly to full time production welding.

During arc time the fume is generated within or close to the worker's breathing zone. Background fume concentrations could also be significant if a large number of welders are working or the work is being performed in a relatively confined space.

Because of the numerous factors that can influence exposure levels during welding, three different types of controls can be used for various welding situations. The controls include: (1) local exhaust ventilation for welding in shops; (2) ambient air cleaning devices to minimize background fume concentrations; and (3) a portable blower for use in confined areas.

Local exhaust ventilation configurations include: a welding bench with a backdraft hood for small to medium work pieces; a fixed close-capture hood placed at the back of a work rest table; a porfootable close-capture system including electrostatic precipitator; or an exhaust hose incorporated into the structure of the welding gun.

Ambient air cleaning devices are designed to lower background welding fumes which escape collection by the local exhaust system. The ambient air cleaner is expected to surpass general dilution ventilation systems in terms of both fume removal and cost.

A portable blower system works by exhausting fumes from a confined space through a large flexible tube.

No commenter questioned technological feasibility for this industry.

SIC 35 - Machinery

In addition to techniques for weld fume control mentioned above, in the manufacture of pumps, employee exposure to welding fumes was controlled (SIC 3561) through the use of an air lux fume eliminator. In the milling of tungsten carbide tools (SIC 354), the placement of local exhaust ventilation controlled cobalt exposures during the transfer of carbide. Oil mist is used in this SIC during the manufacturing of parts on screw machines or other machine parts. There is a wide assortment of engineering and work practice controls to reduce exposure to oil mist [Tr. 8/5/88, p. 7-53]. Since OSHA is retaining the existing PEL for oil mist, OSHA concludes that the PEL is technologically feasible.

In farm equipment manufacturing and repair (SIC 3523), paint mist was controlled through sophisticated application techniques as applied to downdraft spray booths. The use of heated paint in the painting of hay stack wagons allowed the airless atomization to take place at relatively low paint pressures. This resulted in low droplet velocity with little rebound.

In the manufacture of machinery, degreasing operations using refined petroleum solvents are prevalent. The AFL-CIO [Ex 194] and UAW [Ex 197] noted additional feasible measures for control of exposure to refined petroleum solvents (RPS) such as VMP naptha:

(1) Spray application of liquids containing RPS should be permitted only in exhaust ventilated enclosures such as spray booths.

(2) Articles coated with liquids containing RPS should be kept in containers equipped with local exhaust ventilation to prevent evaporation of RPS into work room air.

(3) Equipment for bulk transfer of RPS should be equipped with vapor capture systems.

(4) Exhaust ventilation should not recirculate RPS vapors into workroom air.

(5) Cleaning of floors with RPS should not be permitted. (6) Where spray booths are cleaned with RPS, ventilation should be maintained during cleaning.

(7) Quantities of RPS used and surface area coated with RPS containing liquids should be kept to a minimum.

(8) Splashing of RPS containing liquids or creating of puddles on floor or other surfaces should not be permitted.

(9) Open surface tanks containing RPS should be equipped with covers and local exhaust ventilation. Covers should be closed when not in use. Special attention should be paid to preventing forced expulsion of vapors during addition of materials and entrainment of vapors when articles are added to or removed from open surface tanks.

(10) Open buckets of RPS should not be permitted. Containers for RPS should be equipped with self closing covers. Rags or other material soaking in RPS should be kept in closed containers.

(11) Procedures for response to spills and leaks, including criteria for evacuation of personnel not essential to safe cleanup, should be devised.

(12) Skin contact should be prevented by redesign of operations to eliminate dipping of hands into RPS containing liquids, minimizing splashing or mist contact and wetting of skin and clothing. Gloves and impervious clothing should be supplied where wetting of skin and clothing can't be prevented.

No commenters challenged the technological feasibility of meeting the proposed PELs in this industry.

SIC 36 - Electric and Electronis Equipment

Electric lamp manufacturers (SIC 3641) have reduced mercury vapor in lighting plants. Glass pellets used as starters for fluorescent lamps were flame sealed after mercury had been injected into them. Overhead suction velocity of the exhaust system was increased to reduce mercury overexposure. Also, a special vacuum cleaner was employed to clean the turntable.

The use of styrene for open mold fiberglass operations in the manufacture of household refrigeration equipment (SIC 3632) is similar to the use in SIC 37 -- Transportation Equipment. Thus, respirators may be required to augment engineering controls during manual layup/sprayup operations, as discussed in SIC 37, below.

Technological feasibility was not addressed by commenters to the docket in this sector.

SIC 37 - Transporatation Equipment

The Styrene Information Research Council (SIRC) identified manual layup and sprayup processes as operations in this sector that would not be able to meet either the 50-ppm PEL or the 100-ppm ceiling that the Agency has proposed as limits for styrene [Exs. 187, 3-742; Tr. 3/8/80, p. 5-94]. The open-molding process is primarily used in this sector to make fiberglass boats and fiberglass car and truck bodies (especially bodies for recreational vehicles). The feasibility issue that is raised relates to production operations that involve the spraying of large volumes of resin containing styrene on large surfaces where volatilization occurs [Ex. 198, Tr. 8/3/88, p. 5-95].

The single most extensive data source on exposures to styrene is a study conducted by the State of California's Division of Occupational Safety and Health (DOSH)[Ex. 3-742, Attachment 2]. This study reported the results of an in-depth industrial hygiene survey of styrene and other hazardous exposures in the fiberglass/reinforced plastics industry. A total of 141 workplaces were inspected, and 379 of the 2600 workers employed in these workplaces were sampled over a full workshift [Ex. 3-742, Attachment 2]. The report also recommended the best control measures to minimize hazardous exposures; the focus of the study was on large open-mold sprayup/layup operations because earlier research had shown that these open-molding operations had the highest exposures of all operations in these workplaces [Ex. 3-742, Attachment 2]. Styrene exposures (8-hour TWAs) at these processes ranged from 0.2 to 288 ppm. The overall arithmetic and geometric means for these sample results were 43.0 ppm and 34.0 ppm, respectively [Ex. 3-742, Attachment 2], but exposures in some industries and for some processes were substantially higher.

The range of exposures in boat-building facilities was found to be 3.4 to 90.8 ppm (92 workers sampled); for workers in the recreational vehicle (camper) segment, this range was 7.3 to 130.3 ppm (48 workers sampled). Because few firms in the recreational vehicle and boat-building segments of this industry had adequate and effective ventilation controls, the California study concluded "that feasible engineering controls exist to reduce exposures to levels recommended by ACGIH and NIOSH of [a] 50-ppm TWA and [a] 100-ppm excursion limit for styrene" [Ex. 3-742, Attachment 2, p. 36].

During the course of this rulemaking, OSHA conducted site visits to two boat-making facilities that use the open-mold process to build fiberglass boats [Exs. M-20, M-21]. These two sites were characterized by the industry as a facility that used traditional ventilation to control chemical exposures, and a facility that represented the "best available technology." In both facilities, both the full-shift and the excursion exposures of the gel-coat operators were below the proposed levels. However, the layup and sprayup processes in the traditional facility were conducted in an open area that was ventilated only with general dilution ventilation. In this plant, lamination employees' styrene exposures ranged from 61.9 to 341.5 ppm as 8-hour TWAs and from 98.7 to 311.0 ppm as 15-minute STELs [Ex. M-21]. In the "best available technology" facility, three-sided booths were used for the lamination operations. Here the lamination employees' exposures ranged from 36.7 to 93.8 ppm as 8-hour TWAs, with only one in three exposures below 60 ppm, and from 64.0 to 199.0 ppm as 15-minute STELs [Ex. M-20]. The additional control measures that are available, such as increasing the face velocity of exhaust equipment, may not enable this facility to reduce the exposures of its lamination workers from their current levels (ranging between 36.7 to 93.8 ppm) to levels within the proposed limits without interfering substantially with the correct consistency of the resin.

The extensive exposure and control data reported in the California study indicate that current styrene exposures are within, or can be controlled to, the Agency's proposed limits in some industries and occupations. These data, together with OSHA's on-site observations, are considerably less certain when it comes to the feasibility of the proposed limits for the large-volume open-mold processes necessary to produce boats and campers (as well as other large molded products in other industries).

The California data (Ex. 3-742, Attach. 2] and the OSHA data [Exs. M-20, M-21] showed somewhat different patterns of exposure. Whereas all industry and occupational subgroups of the California data had at least a substantial minority of exposures below 50 ppm, only one exposure observation for lamination in either facility visited by OSHA was below 60 ppm. The maximum 8-hour TWA for boat building observed in the California study was 90.8 ppm, and the large-mold boat manufacturers were described as having low average exposures [Ex. 3-742, Attachment 2]. At the facilities visited by OSHA, on the other hand, the maximum exposure was 341.5 ppm.

The large-scale open-mold processes were described in the California study as "intermittent," and the authors attributed the lower-than-expected styrene exposures to this characteristic [Ex. 3-742, Attachment 2]. SIRC notes that additional feasibility problems may arise in extreme environments, and that northern-tier states, where many boat builders are located, have winter climates that are quite different from that of California [Ex. 181A, p.38]. A report submitted to the docket by the Wisconsin Department of Industry (Ex. HSP) concludes that many of the existing boat-building plants in Wisconsin will be physically unable to accommodate the complex controls needed to reduce employee exposures to styrene to below the 50-ppm TWA and the 100-ppm STEL (Ex. HSP). Since the plants visited by OSHA were in the Midwest, regional differences may help to explain the discrepant findings.

The authors of the Cal/OSHA study qualified their findings as follows:

This study was conducted in a CAL/OSHA compliance mode: This represents two problems: 1) industrial hygienists do not have the luxury of making frequent visits to any one site, and 2) employees have an understandable desire to minimize actual exposures by various means...in order to avoid CAL/OSHA citations and fines [Ex. 3-742, Attachment 2, p. 31].

The California study notes that these factors may well have contributed to the relatively low mean exposures found in the study. However, OSHA notes that industrial hygienists were present in the plants long enough to demonstrate that high exposures can be controlled.

OSHA concludes that, in many operations within SIC 37, the proposed limits for styrene are feasible with the use of engineering and work practice controls. However, the record evidence demonstrates considerable uncertainty about the technical feasibility of achieving the 50 ppm TWA and 100 ppm STEL exclusively by means of engineering controls and work practices during manual sprayup/layup operations in this sector. Accordingly, OSHA concludes that the use of respirators as well as engineering and work practice controls may be necessary to achieve these limits in these manual operations, unless OSHA can demonstrate that engineering and work practice controls alone can achieve the PEL.

OSHA also received some comments regarding the feasibility of achieving the proposed 0.2 ppm PEL for MEKP in boat manufacturing facilities. Robert Schumacker, a certified industrial hygienist representing a group of six manufacturing companies (including the U.S. Marine Corporation), stated that information is lacking as to what concentrations of MEKP currently exist in the workplace, how to measure MEKP in the occupational environment, and the feasibility of engineering controls for reducing exposure to MEKP [Ex. 3-1172, Attachment; Exs. 8-86, 151]. The National Marine Manufacturers Association [Ex. 181] expressed similar opinions.

OSHA believes that the record contains substantial information demonstrating that the final rule's PEL of 0.7 ppm for MEKP is technologically feasible in boat manufacturing facilities. The record contains several NIOSH health hazard evaluations and technical assistance surveys conducted in workplaces where MEKP was used as a reaction catalyst in operations similar to those in boat building, including manual layup and sprayup operations (NIOSH: HE-79-132-673;TA-76-66; and HE-78-3-555). At three sites surveyed, all personal and area samples were below the final rule 0.7 ppm level. The NMMA [Ex. 181] reviewed these and other NIOSH reports (NIOSH: HE-79-092-629; HE-79-012-809) and noted that NIOSH recommended a number of engineering methods to reduce employee exposures to MEKP. These methods, which were supported by NMMA, included:

- Preventive maintenance on a scheduled basis for cleaning and changing filters on spray booths;
- Improvements in general dilution ventilation;
- Release of residual pressure from an MEKP container through the spray gun under local ventilation rather than through the pop-off valve.

OSHA also conducted two site visits to boat-building facilities in which MEKP was used [Exs. 136B]. One plant was a high-volume facility that produced 24 boat perdat, while the other plant produced two to three boats per day. At both of these facilities, MEKP samples taken on gel coat and lamination workers were below the final rule's 0.7 ppm limit.

In regard to sampling and analytical methods for MEKP, OSHA notes that NIOSH has published a method (PECA or 3508) for this substance, and OSHA has developed an in-house method that is available from the Agency on request; the OSHA method was used successfully on the two site visits to MEKP-using facilities conducted for this rulemaking. Therefore, based on the information contained in the record and summarized above, OSHA finds that the 0.7 ppm PEL for MEKP is technologically feasible in transportation facilities.

SIC 38 - Instruments

Many fluxing agents are used in soldering and brazing operations during instrument manufacture. In most cases, these fluxes give off acid or alkali fumes when heated that can irritate the skin. Conducting soldering and brazing operations in well-ventilated areas and use of protective clothing and gloves is recommended.

For many soldering and brazing operations, general dilution ventilation will control fumes and vapors; that is, enough fresh air is added to the contaminated air that hazardous concentrations do not develop.

Local exhaust ventilation is the most effective means of control for airborne contaminants produced by the soldering or brazing process. Local exhaust ventilation can be provided by several types of equipment: freely movable hoods, fixed enclosures (booths), and down-draft benches.

A freely movable hood consists of a movable hood attached to a fan. The fan draws air from the work space and exhausts it outdoors, either directly or through a dust collection system. The hoods are normally constructed so that they can be moved into place by the solderer. The air handling system should move air at least 100 feet per minute across the soldering site at even the most remote point from the exhaust opening. It is important that the exhaust hood be placed as near as possible to the work being done. As such, the proper functioning of a freely movable hood is dependent upon good work practices of the solderer.

In some instances soldering or brazing operations carried out in a fixed location can be provided with a fixed enclosure. This is a structure built around the soldering or brazing operation which has a top and at least two sides. A means for drawing air through the work area is provided so that the work space is flushed continuously with fresh air.

Within such an enclosure, work should be arranged and conducted in such a way that the fresh air enters the enclosure through the worker's breathing zone and then through the work space in which the contaminants are produced. For most fixed enclosures, the air should move at least 100 feet per minute across the entrance to the enclosure.

A third type of local exhaust ventilation system is the down-draft bench or table. The soldering or brazing is performed on a bench or table which has an open grid as the work surface. Air is drawn downward through the grid, into the duct work, and then exhausted.

The Health Industry Manufacturers Association (HIMA) expressed support for "the phased-in period of compliance, which would allow the use of engineering controls, work practices, and respirators for a period of four years while employers evaluate and reduce/eliminate potential exposures to these substances in the workplace" [Ex. 3-910]. They raised no technological feasibility issue.

SIC 39 - Miscellaneous Manufacturing

In the manufacture of hard surface floor coverings (SIC 3996), processes include pre-weighing and blending raw materials, followed by mixing and gelling of the composition in internal batch mixers of the Banbury type or by continuous mixing operations carried out in mixers of the extruder type.

Potential worker exposures may result from dusts of the raw materials as they are handled (automatically or manually) prior to and during charging of the mixer. Fumes and dusts can emanate from leaks on the mixer and from hot, freshly mixed material as it is discharged.

The types of exposures depend on the substances used. Applicable exposure controls include local exhaust ventilation at the mixer doors and over conveyer transfer points. The use of good working practices is extremely effective in controlling exposures during the opening of the mixers and the pouring of materials.

OSHA received one comment related to the issue of technological feasibility in the Sporting and Athletic Goods sector (SIC 3949). Robert Sigler, president of S.R. Smith, Inc., a manufacturer of diving boards, stands, and other reinforced plastics accessories for swimming pools, commented that his plant "will face severe economic hardship and possible closure" if the proposed 50 ppm limit for styrene is retained. Mr. Sigler believes this would be the case because "the entire layout of our manufacturing area and the entire ventilation system would have to be completely stripped and replaced [Ex. 3-380].

OSHA has evaluated the technological feasibility of achieving compliance with a 50 ppm limit for styrene in reinforced plastics operations in several sectors (recreational boat manufacture, cultured marble tubs and showers, and underground storage tanks). Manufacture of fiberglass burial vaults (SIC 3995) is similar to these also. There is a considerable similarity among these various reinforced plastics operations: all involve the use of a styrene resin that is reinforced with fiberglass and applied by "chopper gun" and all involve manual layup and rolling. Thus, although the size and shape of the piece being built may vary, the exposure problems encountered by operators in these facilities are similar in nature. OSHA has determined that employers whose employees perform manual layup and rollup in reinforced plastics operations may need to use a combination of engineering controls, work practices and personal protective equipment to achieve the proposed styrene limit. OSHA's reasoning on this issue can be found in the previous discussion, under SIC 37, Transportation Equipment, and SIC 30, Rubber and Plastic Products.

The Casket Manufacturers Association of America (CMAA) submitted information to the record on the technological feasibility of achieving compliance with the Agency's proposed hardwood dust standard of 1 mg/m(3) [Ex. 8-78]. The CMAA reported that finishing operations, particularly machine and hand sanding of "white" wood caskets, often are associated with dust "levels 3-5 times higher than those in the furniture industry..." [Ex. 8-78)]. In support of this position, the CMAA submitted two sets of exposure results: 9 results from samples taken specifically for this rulemaking, and 24 sample results described as "historical" and drawn from a variety of sources [Ex. 8-78].

Results from the CMAA's recent analysis ranged from 0.8 to 72 mg/m(3);

however sampling times ranged from 90 minutes to 390 minutes. Seven of these nine recent samples showed results below 5 mg/m(3). Results from the historical set of samples ranged from 0.42 to 29 mg/m(3); sampling periods were even shorter than those for the recent set, ranging from 55 to 133 minutes (Ex. 8-78).

The data provided by the CMAA are not adequate to draw firm conclusions about the feasibility of achieving compliance with a 5 mg/m(3) standard for hardwood dust in the hardwood casket manufacturing segment. For example, these data cannot be used to evaluate employees' full-shift exposures because they do not represent 8-hour sampling periods. In addition, no job descriptions or task analyses are presented, and thus no deductions can be drawn about exposures for the unsampled portion of the day. In addition, no details are provided about the specific type of wood involved in casket making, beyond stating that it is a hardwood.

However, OSHA believes that the final rule's 5 mg/m(3) PEL for wood dust is already being achieved in hardwood casket making. The Agency's reasoning is as follows. First, OSHA believes that the results gathered by the CMAA for this rulemaking are more representative than the historical sampling data because they are more recent and generally involved longer sampling periods. Second, OSHA believes that these results reflect wood dust levels during hand or machine finishing operations because it is these operations that the CMAA is concerned about from the technological feasibility perspective. Third, an analysis of these recent results shows that, even using the worst-case assumption that employee exposures continue at the reported levels for the entire work shift (a highly unlikely exposure scenario), 7 out of 9 results would be below 5 mg/m(3) as 8-hour TWAs. For example, the median 8-hour TWA exposure level for this group of samples under this worst-case scenario would be 2.34 mg/m(3).

For these reasons, OSHA finds that hardwood casket manufacturers are already achieving the final rule's PEL of 5 mg/m(3) in almost all cases, even in their dustiest operations (hand and machine finishing). Because Western red cedar is not used to make caskets, the Agency concludes that no casket makers will be affected by the final rule's 2.5 mg/m(3) PEL for this allergenic wood dust. Thus OSHA finds no technological reasons for casket manufacturers to have difficulty complying with the final PEL for wood dust.

SIC 42 - Motor Freight Transportation and Warehousing

Grain elevators whose primary income derives from the storage of grain are classified in SIC 42. Employees working in these elevators have the same kinds of exposures as workers in other types of elevators, which are classified in SIC 51. OSHA's reasoning on the technological feasibility of achieving the proposed grain dust limit in all grain elevators is discussed fully in the technological feasibility section for SIC 51, below.

SICs 40, 45, 47 - Transportation

Cleaning and coating operations are conducted in rail (SIC 40), and air transport industries (SIC 45), as well as in transportation services (SIC 47). These operations require the application of cleaning agents and/or the sandblasting of particles prior to the application of paints or coating. Spraying processes are required for the application of both the cleaning agents and the paints and coatings.

Rail car applications, for example, are generally performed within a large facility, part of which is established as a spray room. The cars are rolled into an enclosed spray area. In manual spray painting rooms, the operator is required to enter and move about the enclosure during spraying. Automatic spray rooms (or booths) are similar but the pressurized spray guns are automatically operated.

Three major spray techniques are used to apply cleaning agents, coatings or paints. These are: compressed air spraying (low-pressure spraying); airless spraying (high-pressure spraying); and electrostatic spraying. The compressed air spray gun atomizes a stream of liquid by impaction with a jet of air. Atomization may take place inside or external to the gun. The air stream and paint droplets intersect the prepared surface. The airless spray gun atomizes the liquid by forcing it through a small orifice under high pressure. The resulting particulate cloud is impelled by the pressure-created momentum toward the surface. Electrostatic spray equipment is based upon the attractive force between two oppositely charged objects. The liquid is atomized by compressed air, airless, or electrostatic techniques. The particles are given either positive or negative charge and the conductive surface to be sprayed is grounded. In general, electrostatic spray techniques result in the lowest exposure levels, followed by airless and then compressed air spraying.

In enclosed spray rooms, particulates enter the operator's breathing zone due to backspray. Exhaust ventilation to control exposure can be designed using down draft or a multiple sidedraft system. Worker positioning in relation to the spray plume is also critical in minimizing exposures. These include minimized line pressure, changing and cleaning of filter banks, enclosure integrity and ventilation maintenance. Personal protective equipment is also generally worn to insure the worker protection.

The industry representatives did not challenge the technological feasibility of the proposed PEL's. However, the Air Transport Association did object to the six month compliance period established by OSHA because of the unique character of the industry and the time required to establish proper controls [Ex. 3-1122]. OSHA recognizes this difficulty but believes that the six month/five year phase in period for complying with this rule, addresses this objection.

SIC 49 - Electric, Gas, and Sanitary Services

Coal-fed power plants present the potential for exposure to coal dust as well as a number of other substances. Coal dust exposures potentially occur in the area where coal is fed into the furnaces. The coal is generally fed into large hoppers off conveyers. Conveyers are filled by front-end loaders from the coal storage area. The operators of the front-end loaders are protected from coal dust exposure with the use of closed, air-conditioned cabs which provide purified breathing air.

Evidence was presented by the Edison Electric Institute regarding technological feasibility during intermittent exposures. Dr. Louis Hosek representing EEI argued that "Many intermittent exposures occur in situations where engineering controls are likely to be substantially less feasible, both technologically and economically, than respirators, personal protective equipment, and work practices" [Tr. 8/11/88, pp. 228, 229).

This would clearly be the case in the tasks of cleaning the boilers and precipitators at electric power plants. Installing engineering controls to reduce exposures inside boilers would not be feasible. A power plant visited for this rulemaking had installed deluge systems for the precipitators to wash down as much fly ash as possible before workers could enter the area for cleaning or maintenance. Workers had to wear protective clothing and respirators when they cleaned the precipitator areas even after the wash system was employed [Ex. 11]. These tasks are occasional, performed maybe four times each year as the opportunity arises when the boilers are shut down. The crews used to perform the cleaning are as large as is practical so that the duration of the operation will be minimized. This is a maintenance situation where engineering controls would not feasibly achieve the exposure level and supplementary respiratory use is appropriate under OSHA's traditional policies.

The Edison Electric Institute questioned whether OSHA had found that compliance with the PEL for ozone was technologically feasible in the electric utility industry [Ex. 133, Tr. 8/11/88, pp. 232-233]. However, the Gulf Power Company, an electric utility, stated that, "Normal operating procedures would prevent exposure exceeding 0.3 ppm, since most operations occur in well ventilated environments" [Exs. 3-938, 3-1144]. Therefore, OSHA concludes that the PEL for ozone is feasible.

SICs 50 and 51 - Wholesale Trade

Some firms in this classification receive liquid chemicals in bulk quantities from a tank truck, store them and then redistribute them in smaller containers. Solvents, for example, emit considerable vapor when poured from one container to another or when a container is being filled, displacing the air in it. Pouring and filling operations are often enclosed to minimize vapor losses (this helps to reduce product loss as well as prevent exposures). In addition, secondary vapor recovery is often incorporated, whereby vapors emitted at the transfer points are captured and returned through a separate circuit to the storage tank from which the volatile liquid is being removed.

Grain dust exposures in this sector occur during grain handling operations in wholesale grain elevators. The majority of commenters on the technological feasibility of the proposed 4 mg/m(3) PEL for grain dust (oats, wheat, and barley) maintained that this limit is not being met currently and cannot be met with available engineering controls [Exs. 8-55, 3-77, 3-201, 3-343, 3-347, 3-496, 3-1119, and 3-1196]. Typical of these comments is one from the Union Equity Coop Exchange, which stated that "over $9 million has been spent to install dust collection equipment in the facilities Union Equity currently operates. Thousands of dollars are spent monthly to maintain and operate this equipment. Many of these systems are state-of-the-art design for functional operation. None of these systems would allow any of our facilities to meet the proposed 4 mg/m(3) exposure level" [Ex. 3-343, p. 2]. Edward X. Junia, Esq., representing the Andersons Management Corporation, was more specific: "there are certain operations in every grain-handling facility where there are no technically feasible engineering controls to reach such a level. The regular unloading and cleaning of storage bins/buildings and the housekeeping activities required under other OSHA standards are two areas where compliance through engineering methods is virtually impossible" [Ex. 3-77, p. 2].

OSHA does not agree that no controls are available to handle employee grain exposures during these operations. For example, in-plant vacuum systems (Farant and Moore, "Dust Exposures in the Canadian Grain Industry," AIHAJ 1978, pp. 177-193) would reduce exposures during housekeeping and maintenance; this control method should be used in lieu of manual sweeping or compressed air cleaning, two housekeeping methods that are still widely used in grain elevators (Ex. 3-751, Attachment, Docket H-0117).

Employers owning elevators that are operated in connection with feed mills (SIC 20) have found that the use of aspirators with filtration systems is highly effective in controlling grain dust during loading and unloading operations in receiving areas (Tr. 8/10/88, p. 10-73). To deal with the problem of grain dust in older mills, owners are replacing old-fashioned wooden legs with "good, tight, enclosed steel legs...old facilities...that had open grain drag conveyers...have been replaced in many cases with enclosed-type conveyers...the conveying systems that used to be open have lids on them...to keep the dust where it belongs" (Tr. 8/10/88, p. 10-80). Such enclosure is recommended by industrial hygienists whenever workers must work in dusty environments.

For some facilities, oil suppression of dust may be a useful control measure. An oil mist, which consists of mineral oil, vegetable oil, or some combination of the two, is normally applied when the grain is received at the mill. Ralph Mourer, testifying for the American Feed Industry Association, stated that oil suppression of dust is a promising control that he has just installed in his feed mills. Although he has not yet had much experience with the system, he noted: "[P]people I've talked [to] and discussed the system with are very pleased" (Tr. 8/10/88, p. 10-78). In an earlier study of grain- handling facilities for OSHA, however, Arthur D. Little, Inc. noted that there are some limitations to this process:

Mineral oil is not approved for use as an additive on food grades of grain by the U.S. Food and Drug Administration. Vegetable oil may be an allowed additive, but its use can cause the grain to adhere into masses in cold climates. Further, there is concern that the oil will become rancid or create a commercially objectionable odor" (Docket H-0117, ADL, p. VI-34).

Scott Bjornsom from Hunter Grain in North Dakota also reported that oil suppression cannot be used for malting barley "because of the absorption with the water in the malt process" (Tr. 8/10/88, p. 10-85). Despite some limitations on its use in elevators, oil suppression appears to be an effective control for many elevators.

OSHA notes that the grain dust exposures of employees working in grain elevators classified in SIC 51 are sometimes below the 10 mg/m(3) (the grain dust limit in the final rule) at the present time. A recent NIOSH study reports that only five percent of samples in the mills surveyed exceeded 10 mg/m(3) (Rankin et al. 1986), and a NIOSH Health Hazard Evaluation from a Cargill elevator showed many sampling results that were below 10 mg/m(3) or only slightly above this level (NIOSH HHE 76-13-316). These exposure levels are being achieved despite the fact that most grain elevators do not now have pneumatic dust control systems (RIA for the Grain Handling standard).

After considering the comments received on the proposed level, the controls available to reduce exposures and the impact on certain segments of the industry, OSHA has set the PEL for grain dust at 10 mg/m(3).

OSHA believes that the controls described above, which are being installed in many elevators at the present time in response to the recent promulgation of OSHA's grain-handling standard, the recommendations of insurers, and the industry's concern for worker safety and health [Tr. 8/10/88, p. 10-73], are capable of achieving the 10 mg/m(3) limit in those facilities and operations that are not now achieving this level. Industry representatives have reported that these systems have several additional benefits for employers; they improve productivity, have a positive effect on the quality of the grain, and create a better working environment [Tr. 8/10/88, pp. 10-80 and 10-81]. Thus, OSHA concludes that a variety of control strategies are available to employers operating grain elevators and these controls are installed in many elevators at the present time. The Agency finds that implementation of these controls will achieve the final rule's grain dust limit of 10 mg/m(3) in those elevators and areas that have not already achieved this level.

SIC 72 - Personal Services

To control dry cleaning emissions (SICs 7216, 7217), louvered wall fans and grilled ducts were installed to provide ventilation. Ceiling exhaust fans provided general ventilation. Natural ventilation was provided by through doors in the production area and by louvered panels along walls in the plant. Forced ventilation was provided by ceiling mounted exhaust fans and evaporative coolers. A local exhaust system with a standard single floor pickup exhausted air through a carbon absorption unit to the outside. Gaskets in machinery doors and ductwork needed routine maintenance to prevent deterioration. Various cleaning machines, pressure filter extractors and dryers were used. Dryers and drying cabinets were provided with local exhaust ventilation.

In addition to the controls mentioned above, information has been reported by the Amalgamated Clothing and Textile Workers Union (ACTWU) which indicates that exposure to perchloroethylene is reduced when using the unitary dry-to-dry equipment (10.7 ppm for operator) as opposed to transfer-type dual washer/dryer equipment (58.4 ppm for operator) [Tr. 8/5/88, pp. 159-186]. ACTWU estimated that over 100,000 workers are exposed to perchloroethylene on a routine basis in the apparel cleaning industry. According to the 1982 Census of Service Industries, there were 13,049 dry cleaning plants in the U.S. that used perchloroethylene, with total employment of 89,896 workers. ACTWU calculated that approximately two-thirds of these workers are exposed in plants using transfer equipment, and one-third of these workers are exposed to perchloroethylene in plants using dry-to-dry equipment [Ex. 8-31]. The ACTWU also commented on the feasibility of reducing the exposure of perchloroethylene to substantially less than the 50 ppm proposed standard. Mitchell Brathwaite, an industrial hygienist representing the ACTWU, stated that OSHA "could reasonably propose a much lower PEL.... For instance, NIOSH reported that [the] mean exposure for 80 percent of [the] plant[s] study [studied] were below 50 ppm. In fact, machine operators in these plants had mean exposures of 22 ppm" [Tr. 8/5/88, p. 190].

NIOSH determined that "the 'combination washer/dryer' machines significantly reduce worker exposure to perchloroethylene when compared to exposures for separate or 'scanter' equipment" [Ex. 150]. Mr. Brathwaite further cited that based on NIOSH data (the Ludwig study) and Mount Sinai Hospital's Division of Occupational Safety and Health Data, "exposures could be reduced below ten ppm" in the dry cleaning industry with the utilization of the dry-to-dry equipment [Tr. 8/5/88, pp. 203-204].

The International Fabricare Institute (IFI) supported the proposed revision of the PEL for perchloroethylene at 50 ppm. They stated that approximately 64 percent of the dry cleaning industry uses dry-to-dry equipment, and that over the past four years about 95 percent of all new equipment sold has been dry-to-dry equipment [Ex. 3-671]. This indicates a continuing increase from 1982 when 35 percent of the firms had dry transfer machines. Based on this trend and the belief that all machines purchased in the future will be of the dry transfer type, all equipment in the dry cleaning industry will be dry-to-dry equipment. This will be accomplished through the normal replacement cycle. The ACTWU [Exs. 153G, 192] reports that the machinery census in Michigan for the period 1983-1988 indicates "there was a 34% increase in dry-to-dry- machines, and an 11% decrease in transfer machines during this period. These data demonstrate vividly that replacement of transfer equipment with dry-to-dry equipment is not only technically feasible, but is indeed the economic choice of employers."

According to the testimony of Mr. William Fisher, vice president of IFI, the ambient perchloroethylene concentration in the cleaning area of a dry cleaning shop is approximately 20 to 30 ppm. The concentration decreases to approximately 10 to 15 ppm in the finishing area (at a range of approximately 15 to 20 feet from the cleaning area) to 1 to 3 ppm at the counter. "There can be some variations in those numbers. However, Ludwig's study from NIOSH indicated the same numbers as did the Westinghouse Behavioral Research Center study..." [Tr. 8/5/88, p. 281]. The actual concentration to which a person would be exposed is dependent upon the ambient environmental conditions and the ventilation characteristics.

Ludwig addressed the issue of engineering controls in dry cleaning facilities:

The dryer is a closed system while in operation and the PCE-laden air leaving the dryer is passed over a water-cooled condenser for solvent recovery before the air is reheated and recirculated through the tumbler.... The processing equipment, whether a combination unit or separate washers and dryers, has interlock systems which insure that there is exhaust ventilation pulling air into the machines and out through ducting whenever the doors are opened. The recommended air velocity in through the loading door is an average of 100 feet per minute across the entire door opening.

An activated charcoal adsorber is often added to the control scheme to remove PCE from the air exhausted from the washer during loading and transfer, and the dryer tumbler when the textiles are being aerated (deodorized), from the air intakes in the processing area, and from the vents of the muck cookers or stills.... The collection efficiency of the activated charcoal is extremely high (greater than 99%) up until breakthrough; however, if the adsorber becomes saturated, all PCE collected by the ventilation system will pass directly through the charcoal. It is for this reason that the adsorber should be vented to the outside of the building.

Local exhaust ventilation in the processing area is also ducted to the adsorber. Ideally, the air intakes are between the level of the equipment doors and the worker's breathing zone. However, due to the mistaken notion that since PCE vapors are heavier than air they collect on the floor, most local exhaust intakes are at floor level; PCE vapors are likely to be found in high concentrations near the floor only if there has been a spill. Along with reducing PCE levels in the processing area, the use of local exhaust when ducted to the adsorber tends to cool the charcoal bed, thereby increasing its adsorption efficiency. Another type of ventilation utilized in some facilities is a low velocity fan 7' to 9' above the floor directed toward the center of the processing area. This concept, when combined with general room ventilation in which the fan is located on the wall or ceiling behind the dry cleaning area, results in reduced employee exposures not only to PCE but also to heat and humidity. A complete room air change every 5 minutes is recommended. Engineering controls such as exhaust ventilation of process equipment vented to a charcoal adsorber, local exhaust in the dry cleaning area, fans, and general room ventilation all contribute to lower ambient PCE concentrations. Also important is an active maintenance program. Typical sources of PCE vapor leaks are the button trap and the doors of the washer, dryer, cooker, or dryer lint trap. Most of these localized leaks are avoided by replacing door gaskets and adjusting the springs and hinges on the doors. Improperly seated air-inlet dampers on the dryer (used during aeration) and ducting are other potential sources of PCE emissions" [Ex. 8-31, Appendix 13].

Data compiled by NIOSH, the dry cleaning industry, the ACTWU, and independent investigators demonstrates that virtually all employers can achieve exposure limits lower than the 50 ppm originally proposed by OSHA by using existing, readily available control technology. OSHA concludes, therefore, that a 25 ppm standard for perchloroethylene is feasible.

SIC 73 - Business Services

Blueprinting and photocopying firms (SIC 7332) control ammonia fumes from blueprint duplication machines through use of local exhaust ventilation. The exhaust system is often built into large, high volume machines. Improvements in work practices control exposures during transfer.

The blueprint reproduction process uses ammonia to develop the image on the finished reproduction. Some machines are designed to contain the ammonia and its vapors; others are vented to the outside atmosphere. However, the odor of ammonia is present around the machines, especially where copies exit the machine and are trimmed [Tr. 8/5/88, p. 216 p. 228, p. 236]. Mr. Lucas Seeman, representing the Association of Reproduction Materials Manufacturers, discussed a survey conducted in 1975 - 1976 of 75 blueprinting materials installations to determine levels of ammonia exposure found during blueprint reproduction. This survey found that, "the dominant part of this group was well under 25 ppm." Mr. Seeman stated that higher levels might occur "at the export end of some of the machines where the paper is coming out" [Tr. 8/5/88, pp. 228, 229]. OSHA concludes that local exhaust would be sufficient to bring these "export end" work stations into compliance with a 35 ppm STEL.

SIC 55, 75 - Automotive Repair Shops, Dealers

Exposure to carbon monoxide presents the major hazard in these industries. To control this in automobile engine reconditioning lines (SIC 7538), exhaust fans and flexible ducts which extend directly over the engines have been installed. OSHA received no comments on the proposed rule from this sector.

SIC 76 - Miscellaneous Repair Services

Many repair services involve welding. In addition to techniques suggested in the discussion on welding and brazing in SICs 34 and 38, another control technique for welding fumes in SIC 7692 uses a "smoke exhaust" welding gun which captures and removes fumes. These guns have some limitations and are applicable to continuous or semicontinuous flux core or metal inert gas welding operations. Crossdraft airflow has also been suggested. The use of a portable fan is not recommended. OSHA concludes that it is feasible to control exposures to the final levels.

There were no docket comments on any aspect of this rulemaking for this sector.

SIC 80 - Health Services

Many medical and dental practitioners perform surgery in outpatient clinics and private offices outfitted for the procedure. Air contamination in an operating room may consist of waste anesthetic, the propellants of different sprays, scrubbing agents, cleansing agents, ethylmethacrylate (released from surgical cement) and the possible decomposition products of the volatile or gaseous agents. The magnitude of gas flow, type of flow circuit and scavenging of waste gases significantly influence the levels of waste gases in the room air. Exposures are usually controlled by general dilution ventilation. Some clinics and offices, which are specifically designed for surgical use, may have local exhaust systems installed.

Glutaraldehyde is used in a limited number of applications, rather than as a general disinfectant. Specific applications include use as a disinfecting agent for respiratory therapy equipment, bronchoscopes, physical therapy whirlpool tubs, surgical instruments, anesthesia equipment parts, x-ray tabletops, dialyzers and dialysis treatment equipment. Presently there are no safer disinfectants which are as effective as glutaraldehyde.

Based on NIOSH Health Hazard Evaluations [1], OSHA concludes that the proposed ceiling limit for glutaraldehyde of 0.8 mg/m(3) (0.2 ppm) is technologically feasible. NIOSH states that those facilities and areas where exposures are below 0.2 ppm achieve these levels through the use of ventilation. NIOSH has found that through a combination of work practice improvements and engineering controls, levels below 0.2 ppm can be achieved. Specific recommendations include using increased dilution ventilation in whirlpool rooms and x-ray rooms, along with the careful application of the glutaraldehyde with a long-handled brush, rather than a spray applicator. NIOSH also recommends the construction of a work station (similar to a lab hood) for cleaning surgical instruments and equipment parts.

OSHA received no comments on the impact of the proposed rule on facilities in this sector.

Personal Protective Equipment

In the operations and processes included in Table F-4 reductions in exposure limits can be achieved through engineering controls and work practice modifications. However, certain generic work activities are more problematical and may require the use of personal protection equipment. OSHA recognizes in 29 CFR 1910.1000 (e), that respiratory protection can be an important adjunct to engineering controls. Because of specific task and process considerations, it may sometimes be necessary to augment engineering controls with the use of respiratory protection.


                               Table F-3

                  INDUSTRIES AND PROCESSES WHERE SKIN
                       PROTECTION HAS BEEN ADDED
________________________________________________________________
SIC | Chemical Name           | Process Name
NO. |                         |
______________________________|__________________________________
    |                         |
25  | N-Butyl Alcohol         | Coating
26  | Methyl Alcohol          | Chemical recovery
27  | Cyclohexanone           | Plate cleaning
    | Furfuryl Alcohol        | Plate cleaning
    | Methyl Alcohol          | Plate cleaning
28  | N-Butyl Alcohol         | Blending, Packaging
    | Carbon Disulfide        | Fiber manufacture
    | Chlorpyrifos            | Blending, Packaging
    | Diazinon                | Blending, Packaging
    | Disulfoton              | Blending, Packaging
    | Hexafluoroacetone       | Reacting
    | Isophorone              | Reacting
    |   Diisocyanate          |
    | Methyl Alcohol          | Blending, Packaging
    | Methyl Parathion        | Blending, Packaging
30  | N-Butyl Alcohol         | Finishing, Trimming, Painting
    | Carbon Disulfide        | Compounding, Mixing and Blending
31  | Methyl Alcohol          | Finishing/Degreasing
    | Triorthocresyl          | Finishing/Degreasing
32  | Methyl Alcohol          | Batch preparation
    | Tin                     | Float Process
34  | N-Butyl Alcohol         | Coating/Painting
35  | N-Butyl Alcohol         | Coating/Painting
36  | N-Butyl Alcohol         | Coating/Painting
    | Mercury                 | Soldering/Brazing
    | Methyl Alcohol          | Cleaning
38  | Mercury                 | Handling of measurement liquids
    |                         | Preparation of Special Tubes
    |                         | Assembling
    | Methyl Alcohol          | Blending/Packaging
39  | N-Butyl Alcohol         | Painting, Coating
45  | N-Butyl Alcohol         | Cleaning/Spraying
    | Methyl Alcohol          | Cleaning/Spraying
55  | N-Butyl Alcohol         | Painting/ Coating
72  | Mercury                 | Embalming
    | Methyl Alcohol          | Embalming
73  | Chlorpyrifos            | Exterminating
    | Diazinon                | Exterminating
    | Dioxathion              | Exterminating
    | Phenothiazine           | Exterminating
75  | N-Butyl Alcohol         | Painting/ Coating
80  | N-Butyl Alcohol         | Disinfectant and solvent use
    | Mercury                 | Preparation of amalgams
    | Sodium Azide            |  Laboratory analysis
________________________________________________________________

                              Table F-4

                       PROCESSES TO BE CONTROLLED

SIC 20 - Food Products

  Refrigeration/charging               Local ventilation:  a hood
                                       exhausted to a baghouse,
                                       appropriate placement of cutoff
                                       values to freezer coils, an alarm
                                       detection system
  Dry ice manufacture and use          Local ventilation:  slotted hood
                                       exhaust system, adjustment of the
                                       number of air changes
  Food storage and preservation        Local ventilation:  slotted hood
                                       exhaust system, adjustment of the
                                       number of air changes
  Grain elevators                      Local ventilation, enclosure of the
                                       Boerner divider

SIC 21 - Tobacco                       Local ventilation

  Cutting and shredding
  Flavor additive blending

SIC 22 - Textile Mills (except 2294)   Local ventilation, pressure failure
                                       alarms for closed systems,
                                       continuous flow indicators to
                                       indicate acceptable airflow
  Wet methods, vacuum cleaning
  Weaving (SICs 2251,2295,2299 only)
  Dying/curing
  Coating/finishing
  Cutting
  Printing
  Spinning (SICs 228, 2299)
  Bonding (SIC 2295)

SIC 2294 - Processed Waste             Local ventilation, pressure failure
                                       alarms for closed systems,
                                       continuous flow indicators to
                                       indicate acceptable airflow
  Wet methods, vacuum cleaning
  Processing of textile mill waste
    and fiber
  Fiber recovery from clippings and rags

SIC 23 - Apparel Products

  Bonding                              Local ventilation, general
  Dying                                ventilation
  Cleaning

SIC 24 - Lumber and Wood Products      Local ventilation:  hoods or
                                       various types of negative pressure
  Drying/baking                        (or combinations of positive and
  Coating/spraying/finishing           negative pressure) devices;
  Sanding/grinding/polishing           enclosed or hooded equipment
  Spraying/coating preservatives       vented to a baghouse, industrial
     (SIC 2491 only)                   vacuum system; enclosures: booth
  Cutting/sawing/planning              or cab supplied with filtered
  Adhesive binding                     conditioned air
  Gluing/hot pressing

SIC 25 - Furniture and Fixtures

  Gluing/hot pressing                  Local ventilation
  Coating/spraying/finishing           Local ventilation: downdraft spray
                                       booths, side draft ventilation;
                                       airless atomizing sprayers,
                                       electrostatic spray
  Spraying/coating preservatives       Local ventilation:  downdraft spray
                                       booths, side draft ventilation;
                                       airless atomizing sprayers,
                                       electrostatic spray guns on
                                       reciprocators; drums equipped with
                                       heavy barrel covers, an internal
                                       agitator, closeable access lines
  Grinding                             Local ventilation
  Sanding/polishing                    Local ventilation
  Deburring                            Local ventilation
  Cutting/sawing/planing               Local ventilation
  Layup/sprayup/coating                Local ventilation
  Baking/drying                        Local ventilation
  Drilling/boring                      Local ventilation

SIC 26 - Paper and Allied Products

SIC 261 - Pulp Mills

  Digester                             Enclosure, local ventilation
  Pulp screening/washing               Ventilation and air purification
                                       in control rooms
  Chemical recovery                    Local ventilation
  Bleaching                            Ventilation and air purification
                                       in control rooms
  Boilers                              Local ventilation
  Water treatment                      Local ventilation, enclosure:
                                       storage of chemicals isolated and
                                       surrounded by dikes, rerouting of
                                       discharge lines
  Recovery/reprocess/reclamation       Local ventilation, enclosure:
                                       storage of chemicals isolated and
                                       surrounded by dikes, rerouting of
                                       discharge lines

SICs 262, 263 - Paper and Paperboard Mills

  Wet end                                Local ventilation,
  Press section                          enclosure:
  Drying                                 air-conditioned cabs or booths
  Size press and coaters
  Calendars and winders


SIC 264, 265, 266 - Paperboard Products and Building Paper

  Mixing/blending (SIC 2641 only)      Local ventilation
  Coating/finishing                    Local ventilation
  Gluing                               Local ventilation
  Drying                               Local ventilation
  Cutting/sawing/planing               Local ventilation
  Packaging                            Local ventilation with partial
                                       enclosure
  Shredding/waste processing           Enclosure, local ventilation
  Stamping/shaping/molding/pressing    Ventilation and air purification in
                                       control rooms

SIC 27 - Printing and Publishing       Local ventilation

  Printing process and plate cleaning
  Platemaking
    Photoengraving
    Gravure
    Lithographic (Offset)
    Screen stencil
    Letterpress
    Flexographic
    Intaglio
  Adhesive binding
  Mono or linotype setting
  Film processing

SIC 28 - Chemicals and Allied Products

  Reaction/fermentation                Local ventilation: enclosing and
                                       exhausting equipment, fitting
                                       vacuum crescents and elephant
                                       trunks on paint sources, fitting
                                       chutes with covers, placing vacuum
                                       attachments on receiving drum
                                       covers, using fixed ductwork as an
                                       exhaust, installing
                                       electronic-spent acid interface
                                       detectors; equipping vessels with
                                       hinged covers
  Separation (many types)              Local ventilation
  Crushing/grinding                    Local ventilation
  Loading/offloading                   Local ventilation with partial
                                       enclosure:  vapor recovery systems
  Drying/baking                        Local ventilation
  Packaging/bagging                    Local ventilation with partial
                                       enclosure:  dust collection hoods
  Reaction/fermentation                Local ventilation
  Coatings/spraying                    Local ventilation: portable hoods
                                       attached to flexible ductwork;
                                       enclosure: vented enclosures kept
                                       under negative pressure by a
                                       ventilation system
  Blending/mixing/formulating          Local ventilation: enclosing and
                                       exhausting equipment, modification
                                       of hoods, fitting vacuum crescents
                                       and elephant trunks on paint
                                       sources, fitting chutes with
                                       covers, placing vacuum attachments
                                       on receiving drum covers, using
                                       fixed ductwork as an exhaust,
                                       installing electronic-spent acid
                                       interface detectors; equipping
                                       vessels with hinged covers
  Impregnation                         Local ventilation
  Extrusion                            Local ventilation
  Recovery/reprocess/reclamation       Enclosure, local ventilation

SIC 29 - Petroleum Refining

  Coke production                      Worker enclosure, scrubber,
                                       computer control instrumentation,
                                       hardware modifications
  Blending/mixing                      Local ventilation

SIC 2911 - Petroleum Refining

  Loading and unloading                Local ventilation with partial
                                       enclosure
  Sampling                             Local ventilation with partial
                                       enclosure: sampling boxes that vent
                                       gases and vapors away from operator
                                       and/or shield the operator from
                                       accidentally splashed or spilled
                                       material
  Process inspection and
    supervision                        Local ventilation with partial
                                       enclosure
  Quality control analysis             Local ventilation with partial
                                       enclosure: exhaust fans and
                                       ventilation hoods
  Waste water treatment                Enclosure, local ventilation

  Batch process coke production        Worker enclosure, scrubber,
    and removal                        computer control instrumentation,
                                       hardware modifications

SIC 2951 - Paving Mixtures

  Materials receiving and handling     Local ventilation with partial
                                       enclosure
  Measurement                          Local ventilation
  Drying/baking                        Local ventilation
  Mixing (Continuous/Batch)            Local ventilation

SIC 299 - Miscellaneous Petroleum and Coal Products

  Materials receiving and handling     Local ventilation with partial
                                       enclosure
  Blending/mixing                      Local ventilation
  Reprocessing or reclamation          Enclosure, local ventilation
  Packing and loading                  Local ventilation with partial
                                       enclosure
  Adhesive binding                     Local ventilation

SIC 30 - Rubber and Miscellaneous      Local ventilation:  hoods,
         Plastics Products             automated batching systems, use of
                                       rubber bins rather than screw
                                       conveyers, substitution of
                                       chemicals

SIC 301 - Tires and Inner Tubes

  Materials receiving and              Local ventilation with partial
    initial handling                   enclosure
  Compounding and mixing               Local ventilation: hoods; automated
                                       batching systems, use of rubber
                                       bins rather than screw conveyers,
                                       substitution of chemicals
  Vulcanization or curing              Local ventilation
  Calendering and milling              Local ventilation
  Solvent mixing and distribution      Local ventilation
  Tire building                        Local ventilation
  Reblending/remixing                  Local ventilation
  Coating/spraying                     Local ventilation
  Stamping/shaping/molding/pressing    Local ventilation

SIC 306 - Fabricated Rubber Products

  Materials receiving and              Local ventilation with partial
    initial handling                   enclosure
  Blending, compounding, and mixing    Local ventilation: hoods, automated
                                       batching systems, use of rubber
                                       bins rather than screw conveyers,
                                       substitution of chemicals
  Extrusion                            Local ventilation
  Coating/spraying                     Local ventilation
  Calendering and milling              Local ventilation
  Vulcanization or curing              Local ventilation
  Finishing, trimming, and painting    Local ventilation

SIC 307 - Miscellaneous Plastic Products

  Material handling                    Local ventilation enclosure
  Blending, mixing and compounding     Local ventilation: hoods, automated
                                       batching systems, use of rubber
                                       bins rather than screw conveyers,
                                       substitution of chemicals
  Calendering                          Local ventilation
  Molding and mold cleaning            Local ventilation
  Assembly (including lamination,
    gluing, etc.)                      Local ventilation
  Foam processing                      Local ventilation
  Finishing, trimming, and painting    Local ventilation
  Coating/spraying                     Local ventilation

SIC 31 - Leather and Leather Products (except SIC 311)

  Gluing and cementing                 Local ventilation
                                       Work practice changes


 SIC 311 - Leather Tanning and Finishing

   Preservation                        General ventilation
   Defestation and disinfection        General ventilation
   Beamhouse                           Local ventilation
   Tanning                             Local ventilation
   Splitting and shaving               General ventilation
   Neutralizing                        Local ventilation
   Retaining                           Local ventilation
   Coloring or dyeing                  General ventilation
   Fat liquoring                       General ventilation
   Drying                              Local ventilation
   Finishing (includes degreasing)     Local ventilation

SIC 32 - Stone, Clay, and Glass Products

SICs 321, 322, 323 - Glass

  Batching                             Local ventilation: industrial
                                       vacuums system/work practice
                                       changes
  Melting                              Local ventilation
  Plate process (SIC 3211 only)        Enclosure
  Sheet process (SIC 3211 only)        Enclosure
  Float process (Sic 3211 only)        Local ventilation
  Molding and blowing                  Enclosure
  Annealing                            Local ventilation
  Coating and etching                  Local ventilation

SICs 324, 327 - Cement, Concrete, Gypsum, Plaster

  Gypsum, Plaster                      Local ventilation:  local exhaust
  Crusher, grinder and sizing          tubes, use of wet materials,
  Blending                             industrial vacuum systems
  Calcining kiln

SIC 325, 326 - Clay, Pottery

  Crushing, grinding, calcining        Local ventilation:  local exhaust
  Slip house (blending)                tubes, use of wet materials,
  Forming and shaping                  industrial vacuum systems
  Biscuit firing
  Glaze application
  Gloss firing
  Decoration

SIC 328 - Stone

  Drilling, cutting, flame-jet         Local ventilation
    lancing
  Chipping and grinding
  Surface polishing

SIC 329 - Abrasives

  Crusher, grinding, sizing            Local ventilation
  Calcining (abrasives)                Local ventilation
  Bonding                              Local ventilation
  Melting (Cupola furnace) and raw     Local ventilation
    material handling
  Fiber forming (steam jet process,    Enclosure
    Powell process, Downey process,
    dry spinning)
  Blowing/molding                      Enclosure

SIC 33 - Primary Metal Industries

SIC 331 - Basic Steel Products

  Coke Manufacture                     Worker enclosure, scrubber,
                                       computer control instrumentation,
                                       hardware modification, enclosed
                                       vessels and process equipment
  Ore Handling                         Local ventilation
  Blast furnace operation              Local ventilation; vented to an
    (including furnace charging        electrostatic precipitator and/or
    BOF, ladle repair)                 baghouse, covering of runners;
                                       replacing older blast furnaces;
                                       filtered air for work enclosure
  Melting, pouring (electric arc or    Local ventilation: roof-level hoods
    induction) (including electrode    and ducts, ventilation to an
    production/baking)                 electrostatic precipitator or
                                       baghouse
  Hot shaping of metal (including      Local ventilation
    rolling mill, metal extrusion,
    wire drawing, forging press)
  Annealing, quench and temper         Local ventilation
  Pickling                             Local ventilation
  Hot dip galvanizing                  Local ventilation
  Cold rolling mill                    Local ventilation
  Abrasive blasting                    Local ventilation
  Grinding/polishing                   Local ventilation
  Degreasing                           Local ventilation
  Sintering                            Local ventilation: hoods and sinter
                                       air cooler directed to a baghouse
  Hot strip                            General dilution ventilation
                                       enclosure: air-conditioned control
                                       stations

SIC  332 - Iron and Steel Foundries

  Melting (electric arc or cupola      Local ventilation:  overhead canopy
    (including electrode production    hoods, ventilation of booths,
    and baking, charging and           mancooler fans enclosed crane
    ladle arc repair)                  operator cabs
  Metal (sand) casting (or pouring)    Local ventilation
  Investment casting                   Local ventilation
  Annealing, quench and temper         Local ventilation
  Abrasive blasting                    Local ventilation
  Finishing (including: torch
    cutting, grinding/polishing)       Local ventilation
  Degreasing                           Local ventilation
  Shakeout                             Local ventilation
  Coremaking                           Local ventilation:  to provide
                                       negative pressure at the core box
  Moldmaking                           Local ventilation
  Sand reclamation                     Local ventilation, mechanical
                                       shakeout and automatic sand
                                       handling complete with dust
                                       collection, make-up air systems

SIC 333 - Primary Nonferrous Metals

  Ore handling                         Local ventilation: covers, hoods,
                                       and exhaust systems for belts,
                                       material handling, and transfer
                                       systems, enclosing and exhausting
                                       equipment; enclosed air-conditioned
                                       control booth; computer controlled
                                       systems; movable nozzles; wet
                                       techniques in storage; general
                                       dilution ventilation; replacing the
                                       dross handling operation with a
                                       dross mill
  Melting (electric arc or induction)  Local ventilation:  slotted hoods,
    including electrode production     secondary converter system;
    /baking, charging, ladle repair    induction furnaces; enclosed
    demagging for Alum. plants only)   consoles; operating electric
                                       furnaces of negative pressure
  Metal pouring                        Local ventilation: hooding with
                                       air-operated doors, a dual draft
                                       system, an exhaust system vented to
                                       a dry scrubber;
  Hot shaping (including rolling       Local ventilation
    mill, forging, wire drawing)
  Annealing, quench and temper         Local ventilation
  Degreasing                           Local ventilation

SIC 334 - Secondary Nonferrous Metals

  Melting (electric are or induction)  Local ventilation:  slotted hoods,
    including electrode                secondary converter hoods,
    production/baking, charging,       converter gas handling system,
    ladle repair, demagging (for       induction furnaces; enclosed
    Alum. plants only)                 consoles; operating electric
                                       furnaces at negative pressure
  Metal casting or pouring             Local ventilation
  Forging press (SIC 334, only)        Local ventilation
  Torch cutting                        Local ventilation
  Raw materials preparation            Local ventilation: industrial
   (SIC 334, only)                     vacuum systems, pneumatic aerators
    [including metal preheat           use of deadbeds, eliminate air
     borings dryer, scrap              lancing.
     shredder, slag recovery]
  Degreasing                           Local ventilation


SIC 335 - Nonferrous Rolling and Drawing

  Hot shaping (including rolling       Local ventilation
    mill, wire drawing, metal
    extrusion)

SIC 336 - Nonferrous Foundries

  Melting (electric arc or induction)  Local ventilation
    (including electrode production
    /baking, charging, ladle repair)
  Metal (sand) casting (or pouring)    Local ventilation:  hooded
                                       enclosures, flexible ducting
                                       connecting a mobile hood to a
                                       traveling exhaust carriage
  Investment casting                   Local ventilation
  Annealing, quench and tempter        Local ventilation
  Pickling                             Local ventilation
  Abrasive blasting                    Local ventilation
  Grinding/polishing                   Local ventilation
  Degreasing                           Local ventilation
  Shakeout                             Local ventilation
  Coremaking                           Local ventilation
  Moldmaking                           Local ventilation
  Sand reclamation                     Local ventilation

SIC 339 - Miscellaneous Primary Metal Products

  Sintering                            Enclosure; local ventilation: use
                                       of closed screw conveyers, baghouse
                                       and electrostatic precipitators
  Strip annealing                      Local ventilation: use of closed
                                       screw conveyers, baghouses, and
                                       electrostatic precipitators
  Bimetal production                   Local ventilation: use of closed
                                       screw conveyers, baghouse, and
                                       electrostatic precipitators

SIC 34 - Fabricated Metal Products

  Electroplating                       Local ventilation

SIC 341, 342, 343, 348 - Cans, Cutlery, Hand Tools
        Heating Equipment Ordinance

  Pressing                             Local ventilation
  Acid washing                         Local ventilation
  Degreasing                           Local ventilation
  Painting and coating                 Local ventilation
  Electroplating                       Local ventilation
  Welding                              Local ventilation: a welding bench
                                       with a backdraft hood, a fixed
                                       close-capture hood placed at the
                                       back of the work table, a portable
                                       close-capture system including an
                                       electrostatic precipitator, an
                                       exhaust hose incorporated into the
                                       structure of the welding gun;
                                       ambient air cleaning devices; a
                                       portable blower for use in confined
                                       areas
  Grinding/polishing                   Local ventilation
  Abrasive blasting                    Local ventilation
  Hot shaping (including: rolling      Local ventilation
    mill, wire drawing, metal
    extrusion)

SIC 344 - Structural Products

  Painting and Coating                 Local ventilation
  Welding                              Local ventilation; welding bench
                                       with a backdraft hood, a fixed
                                       close-capture hood placed at the
                                       back of the work table, a portable
                                       close-capture system including an
                                       electrostatic precipitator, an
                                       exhaust hose incorporated into the
                                       structure of the welding gun;
                                       ambient air cleaning devices; a
                                       portable blower for use in confined
                                       areas
  Grinding/polishing                   Local ventilation
  Abrasive blasting                    Local ventilation
  Acid washing                         Local ventilation

SIC 345, 347 - Screw Machine Products

  Coating and engraving                Local ventilation
  Coating (enamels, lacquers,          Local ventilation
    varnishes) (SIC 347 only)
  Hot dip galvanizing (SIC 347 only)   Local ventilation: two-sided
                                       lateral exhaust system,
                                       two-sided slot ventilation system
                                       a cover which is hinged to a
                                       ventilation manifold
  Engraving and etching (SIC 347 only) Local ventilation
  Degreasing                           Local ventilation
  Grinding/polishing                   Local ventilation
  Hot shaping (SIC 345 only)           Local ventilation
  Acid washing                         Local ventilation

SIC 346 - Iron and Steel Forgings      Local ventilation

  Hot shaping
  Acid washing
  Pressing

SIC 35 - Machinery                     Local ventilation

  Pressing                             Local ventilation
  Acid washing                         Local ventilation
  Degreasing                           Local ventilation
  Painting and coating                 Local ventilation: downdraft spray
                                       booths
  Electroplating                       Local ventilation
  Grinding/polishing                   Local ventilation
  Abrasive blasting                    Local ventilation
  Welding                              Local ventilation; welding bench
                                       with a backdraft hood, a fixed
                                       close-capture hood placed at the
                                       back of the work table, a portable
                                       close-capture system including an
                                       electrostatic precipitator, an
                                       exhaust hose incorporated into the
                                       structure of the welding gun;
                                       ambient air cleaning devices; a
                                       portable blower for use in confined
                                       areas; an air lux fume eliminator
  Hot shaping (including: rolling      Local ventilation
    mill, wire drawing, metal
    extrusion)
  Soldering (SIC 357, only)            Local ventilation
  Refrigerant charging (SIC 358, only) Local ventilation

SIC 36 - Electric and Electronic       Local ventilation
           Equipment

  Cleaning

SICs 361, 362, 363 - Transmission      Local ventilation
  Distribution; Industrial Household

  Pressing
  Acid washing
  Degreasing
  Painting and coating
  Refrigerant cooling
  Electroplating
  Grinding/polishing
  Abrasive blasting
  Welding
  Soldering
  Epoxy coating

SIC 364 - Lighting and Wiring          Local ventilation

  Wire drawing                         Local ventilation
  Patenting                            Local ventilation
  Descaling                            Local ventilation
  Coating                              Local ventilation
  Extrusion                            Local ventilation
  Cleaning                             Local ventilation
  Coil production                      Local ventilation
  Coating and drawing                  Local ventilation
  Gas filling                          Local ventilation: increase
                                       overhead suction velocity,
                                       industrial vacuum systems
  Glass blowing                        Local ventilation
  Soldering                            Local ventilation
  Glassmaking                          Local ventilation


SIC 365, 367, - Radio, TV:             Local ventilation
         Communications; Electronics

  Semiconductor-photoresist stripping
  Semiconductor-chemical etchants
  Semiconductor-diffusion and ion implant
  PC-boards-etching
  PC-boards-soldering
  Mixing of ceramic powders

SIC 369 - Miscellaneous Electrical     Local ventilation

  Ingredients grinding
  Mixing
  Casting
  Assembly

SIC 37 - Transportation Equipment      Local ventilation

  Metal melting                        Local ventilation
  Metal pouring                        Local ventilation
  Hot metal working (rolling, shaping  Local ventilation
    or drawing)
  Metal machining or grinding          Local ventilation: hoods, exhaust
                                       fans, "upblast" roof ventilator
                                       fans to change airflow
  Welding or brazing                   Local ventilation; a four sided
                                       enclosure with electrostatic
                                       precipitator ventilation
  Solvent or vapor degreasing          Local ventilation
  Painting or coating                  Local ventilation: enclosed booths,
                                       three-sided booths for lamination
  Degreasing/cleaning                  Local ventilation
  Electroplating or electrical
    discharge machinery                Local ventilation

SIC 38 - Instruments                   Local ventilation

  Forming/fabricating of metal         Local ventilation
  Welding                              Local ventilation
  Injection molding                    Local ventilation
  Handling of measurement and testing
     liquids, gases, materials         Local ventilation
  Quality control testing              Local ventilation
  Foaming, packaging                   Local ventilation
  Coating, painting                    Local ventilation
  Sterilization                        Local ventilation
  Film and spring papermaking,         Local ventilation
    coating (SIC 3861 only)

SIC 39 - Miscellaneous Manufacturing Industries

  Roughing milling or sawing           Local ventilation
  Sanding                              Local ventilation
  Gluing                               Local ventilation
  Finishing or staining                Local ventilation
  Welding, casting, brazing            Local ventilation
  Hot metal work                       Local ventilation
  Mono or linotype setting             Local ventilation
  Abrasive blasting                    Local ventilation
  Degreasing                           Local ventilation
  Electroplating                       Local ventilation
  Machining                            Local ventilation
  Blending, mixing or compounding      Local ventilation
  Molding or mold cleaning             Ventilation and air purification
                                       in control rooms
  Foam processing                      Local ventilation
  Painting/cooling                     Local ventilation
  Metal melting and pouring            Local ventilation
  Pressing                             Local ventilation
  Stamping/shaping/molding/pressing    Ventilation and air purification
                                       in control rooms
  Cutting/sawing/planning              Local ventilation
  Lacquering/enameling                 Local ventilation
  Bristle/fiber cleaning               General ventilation
  Metal plating                        Local ventilation
  Engraving/etching                    Local ventilation
  Acid washing                         Local ventilation
  Hot dip galvanizing                  Local ventilation
    welding, degreasing, metal working
    sand blasting
  Engine fueling                       General ventilation
  Handling spills, leaks               Local ventilation

SIC 45 - Transportation by Air

  Loading/offloading                   Local ventilation
  Maintenance-related activities:      Local ventilation: spray room
    Cleaning/coating/spraying
    welding, degreasing, metal working,
    sand blasting
  Engine fueling                       General ventilation
  Handling spills, leaks               Local ventilation
  Fuel preparation                     Local ventilation with partial
                                       enclosure
  Deicing                              Local ventilation
  Refueling                            General ventilation
  Painting/coating                     Local ventilation

SIC 47 - Transportation Services

  Loading/offloading (SIC 4742 only)   Local ventilation with partial
                                       enclosure
  Maintenance related activities       Local ventilation: spray room
  Engine fueling and fumes             General ventilation
  Handling spills, leaks               Local ventilation
  Special care of lading service       General ventilation

SIC 49 - Electric, Gas, and Sanitary Services

  Maintenance related activities       Local ventilation
  Boiler furnace feed                  Enclosure, local ventilation
  Stripping of chemicals               Local ventilation
  Collection/transport                 Respirators
  Engine fueling                       General ventilation
  Odorant addition                     Local ventilation
  Condensate collection                Local ventilation
  Incineration (SIC 4953, only)        Local ventilation
  Detoxification (SIC 4953, only)      Local ventilation
  Recycling, reclamation (SIC 4953     Enclosure, local ventilation
    only)
  Chemical preparation/application     Local ventilation
  Sampling of pipelines                Local ventilation with partial
                                       enclosure
  Water purification                   Enclosure, local ventilation
  Water treatment                      Enclosure, local ventilation

SIC 50 and 51 (except 5093) - Wholesale Trade

  Material handling, shipping,         Local ventilation with partial
    receiving                          enclosure, secondary vapor recovery
  Material packing or repacking        Local ventilation
  Grain elevators                      Local ventilation

SIC 5093 - Scrap and Waste Materials

  Assembling and collecting scrap and  Respirators
        waste materials
  Breaking up waste materials          Enclosure, local ventilation
  Sorting scrap and waste materials    Enclosure, local ventilation
  Baling or compacting                 Enclosure, local ventilation

SIC 72 - Personal Services

  Washing (SIC 721 only)               General ventilation
  Dry cleaning (SIC 721 only)          Local ventilation: equipment
                                       change; exhaust ventilation of
                                       process equipment vented to
                                       charcoal adsorber, louvered wall
                                       fans and grilled ducts, louvered
                                       wall panels, evaporative coolers;
                                       general ventilation ceiling exhaust
                                       fans
  Manicure/pedicures (SICs 723
    and 724)                           General ventilation
  Permanents (SICs 723 and 724 only)   General ventilation
  Coloring (SICs 723 and 724 only)     General ventilation
  Embalming (SIC 726 only)             General ventilation

SIC 73 - Business Services  Local ventilation

  Blueprint copying (SIC 733 only)
  Exterminating (SIC 734 only)
  Photofinishing (SIC 739 only)

SIC 55 75 - Automotive Repair Shops, Dealers

  Confined space - exhaust fume        General ventilation: exhaust fans
                                       and flexible ducts
  Welding                              Local ventilation
  Paint stripping                      Local ventilation
  Cleaning with solvents               Local ventilation
  Painting/coating                     Local ventilation
  Other solvent use                    Local ventilation

SIC 76 - Miscellaneous Repair Services Local ventilation

  Welding or brazing                   Local ventilation: "smoke exhaust"
                                       welding gun, crossdraft airflow
  Paint stripping                      Local ventilation
  Sanding or grinding
  Gluing                               Local ventilation
  Painting, coating or lacquering      Local ventilation
  Other solvent use                    Local ventilation

SIC 80 - Health Services               Local ventilation, general dilution
                                       ventilation

  Administration of anesthesia         Local ventilation
  Preparation of dental amalgams       Local ventilation
    and alloys
  Laboratory procedures such as        Local ventilation
    tissue staining
  X-ray film processing                Local ventilation
  Use of disinfectants, solvents       Local ventilation, respirators
  Making of dental appliances          Local ventilation

Examples of these are:

Exterminating: Exposure of pesticide applicators cannot be controlled through engineering controls because their work does not take place in a fixed place of employment, but rather at a customer's facility. Personal protective equipment and/or work practice controls would therefore be required. EPA has jurisdiction in most situations.

Welding: In certain situations, such as in confined spaces, or where the welder must be positioned directly above the fume plume, welders cannot be sufficiently protected by local exhaust ventilation. Personal protective equipment would be required.

Maintenance Activities in all SIC codes. In certain cases, it may be more difficult to control exposures of plant maintenance personnel by engineering controls. These maintenance employees may work in areas not normally covered by engineering controls or in situations where engineering controls must be shut down. Respiratory protection is therefore sometimes the appropriate control technology.

Painting and Coating Activities in all SIC codes. Although production spray painting operations are performed in exhausted paint booths, the painting of many larger non-production items, such as construction equipment and heavy machinery, requires that the operator enter the booth. The booth is then primarily a control to prevent migration of the paint spray into other areas of the plant. In these circumstances it is usually necessary to provide respiratory protection to the workers painting.

In addition to these general industry operations, certain industry specific situations have been identified where the use of respirators is recognized as an important complement to other control measures. These situations include the following:

* During exposure to carbon disulfide in the cellulosic food casings industry.

* During exposure to carbon disulfide while changing spinerettes, removing filiment bundles and making product line changes in the manufacture of rayon fibers.

* During episodic emissions of sulfur dioxide in the smelting of copper. * During manual layup/sprayup operations using styrene. * During episodic emissions or intermittent worker exposures of such fumes as carbon monoxide or sulfur dioxide from blast furnaces or BOF.

In addition to the above examples, a number of the substances included in this rulemaking carry the designation "Skin." This refers to potential exposure through the skin. Table F-3 presents a list of chemicals for which skin protection would be required. Employees exposed to substances with the "Skin" notation would be required to wear protective equipment, including gloves, long sleeved shirts and coveralls.

Products are commercially available to adequately protect workers from dermal exposure. In some cases the permeability of currently used materials may be inadequate and firms will have to change the specific product now used to one offering greater protection.

G. Costs of Compliance

Costs of compliance result from the purchase, installation, operation and maintenance of equipment to maintain workers' exposures at or below the levels specified in the final standard. Costs are related to the engineering controls and personal protective equipment needed for specific processes which involve the use of hazardous substances. Given the large number of substances being regulated, the cost assessment was required to examine a large number of processes over many industry segments. The approach needed to be generic in scope and specific in detail. OSHA has reviewed this approach and the resulting cost analysis and incorporated extensive public testimony and voluminous docket submissions. The Agency concludes that the costs presented in this chapter accurately reflect industries' requirements for compliance.

Existing data sources and expert judgment were initially used to sort the approximately 430 substances being regulated, by industry and by process within industry segments. Given the large number of substances being regulated, a process orientation rather than a chemical-specific focus was recommended, since prescribed engineering controls can address worker exposure problems to several chemicals, involving the same general process, simultaneously. The approach has proven to be efficient analytically and reduces the problem of double counting the costs of similar or the same engineering controls for separate chemicals involved in the same process or operation.

OSHA had a large amount of exposure data in its Integrated Management Information System (IMIS) and from NIOSH and other sources. But to improve the available information on the use of substances, OSHA decided to engage in a nationwide field survey of affected establishments. This survey, involving about 5,700 establishments in both manufacturing and non-manufacturing sectors, has provided valuable information on chemical usage by industry process and potential worker exposures to these chemicals. Supplement 1 contains a description of the sample survey design and a statistical evaluation of the data collected.

In order to maximize the efficiency of this nationwide sample survey and limit the number of required sample observations per SIC category, a considerable effort was made to verify chemical by industry usage from existing data sources and to make best estimates of where likely or potential worker exposure problems (and consequently engineering costs) existed. For the purposes of the statistical survey being conducted, the larger the suspected potential exposure/cost problem in a particular industry sector, the more important it was to insure a large enough sample of firms in that sector so as to reduce the standard error of the cost estimates.

The following sections of this chapter outline the methodology adopted to identify:

* Chemicals by their industrial usage and employee exposures * Processes involving known or suspected chemical exposures and control costs * Industry costs for the controls needed to reduce industry exposure levels.

Linking Hazardous Substances by Industry Use and Employee Exposure

Figure G-1 presents a flow chart of the methodology used for identifying chemicals by industry use and employee exposure. The first step in the methodology was an analysis of the chemicals for which OSHA proposes new exposure limits. The 1982 NIOSH National Occupational Exposure Survey (NOES) and the OSHA IMIS data files were searched to determine the potential for worker exposure to each of the chemicals on the proposed list. The objective of this analysis was to create a subset of chemicals which are known to be present in specific industries at exposure levels above the proposed limits. These chemicals would then be considered to generate potential compliance costs within a specific industry sector.


  Figure G-1. Methodology for Identifying Chemicals With Potential Cost in
              Each 4-Digit SIC.

(For Figure G-1, Click Here)

The 1982 NIOSH National Occupational Exposure Survey (NOES) data (supplemented by results from NIOSH's 1972 survey) provided an estimate of the number of workers potentially exposed to a specific chemical in a four-digit SIC. OSHA divided this estimate by the total number of employees in an industry segment to get a percentage of workers potentially exposed to that chemical. If 5 percent or more of the workers were potentially exposed, that chemical was considered to present a potential cost within the four-digit SIC. For example, in SIC 3011, Tires and Innertubes Manufacturing, 1,532 persons were potentially at risk of exposure to n-hexane at the sample of plants included in the NOES database. This represented 21.9 percent of all workers sampled in the four-digit industry sector and this chemical would have a potential cost impact depending upon current exposure levels.

From the OSHA IMIS data, the severity of exposure within a four-digit SIC was estimated. OSHA compared the total number of monitored readings for each chemical with the number of readings which exceeded the proposed limits and calculated the percentage of all sample monitor readings which were above the proposed limits. If there were no readings which exceeded the proposed limits, the chemical was not considered to have a compliance cost within the four-digit SIC. If 5 percent or more of the readings exceeded the proposed limits, then the chemical was identified as having a potential compliance cost within the four-digit SIC. For example, in SIC 2641, Paper Coating and Glazing, 22 samples were taken for n-hexane. Thirteen of these, or 59 percent, were above the proposed standard for n-hexane. This chemical, therefore, was believed to have a potential cost impact and questions regarding its use were included in the field survey. Chemicals with non-compliance percentages between zero and 5 percent were evaluated individually by industrial hygienists to determine whether or not specific survey questions needed to be asked about their industrial usage.

In addition to the IMIS and NOES databases, a survey of about one dozen industrial hygienists was conducted. The purpose of this survey was to identify any additional hazardous substances or industry sectors not identified in the IMIS or NOES databases with potential exposure problems at new recommended levels. For example, in SIC 2891, Adhesives and Sealants Manufacturing, the surveyed industrial hygienists reported that n-hexane overexposures could exist under the proposed standard. (Overexposures in SIC 2891 were not previously identified in the IMIS or NOES databases.) The information from all sources was combined to compile a preliminary list of substances with potential compliance costs by four-digit SIC classification. To further refine the list of chemicals, a second group of six industrial hygienists, using personal industry knowledge and the information gathered from the survey of the initial group of industrial hygienists, reviewed once again the chemicals which appeared in the NOES and IMIS datasets. They also made chemical by industry use linkages when particular chemicals were known to be present in certain SICs, but had not been identified in the NOES and IMIS databases.

Upon completion of the two-tier industrial hygienist review, a list of chemicals believed to be present at exposure levels above the proposed standard, within specific four-digit SIC industry sectors was finalized. This list identified those industry segments where potential compliance costs would be incurred to achieve the proposed standards. The presence of the identified chemicals was confirmed by survey respondents. The method used during the survey listed likely chemicals and asked for any other chemicals present.

Industrial Processes and Control Costs

The number of industrial processes, exposure levels, and exposure controls in place varies greatly within industry segments. In order to efficiently structure the statistical sample of surveyed firms, it was necessary to make a best estimate of which industry segments were likely to experience compliance costs. As noted above, the survey was designed to limit the standard error for potential high cost industry sectors. To concentrate the survey on the potential high cost sectors, a process orientation was adopted which supplemented and refined the chemical use information. The validity of this approach was confirmed in the review of docket materials. The vast majority of submissions that addressed industry costs linked process operations with compliance costs. Industry sectors having few processes and chemicals and low potential exposure levels (and consequently low potential compliance costs) were included in OSHA's secondary data collection and evaluated by experts, but not included in the sample survey.

A team of engineers and industrial hygienists analyzed each four-digit SIC to assess the processes in which worker exposure to listed chemicals occur. Examples of industrial processes included grinding, mixing, spraying, degreasing, separation, bagging and loading. A list of potential cost chemicals and related processes was then developed to identify potentially high impact (cost) industries. The presence of the identified processes was confirmed by survey respondents. The method used during the survey listed likely processes and asked for other processes ongoing at the establishment. In general, an industry segment with a relatively large number of processes using chemicals with suspected high exposure levels was sampled at the three-digit industry level. Industries with fewer processes and low chemical exposures were sampled at the two-digit level. (See Supplement 1 for a more detailed explanation of the survey design.) Approximately 5,700 respondents in the survey were asked to verify the chemicals used, manufactured or generated by process within their establishment. Thus, chemicals were linked to specific processes, process controls and workers exposed at the process in the surveyed industries. Control methods and costs were then assigned for each process where employee exposures would exceed the proposed PELs.

Controls were assigned to protect workers exposed to all chemicals in total at a process. The controls were designed and costed to lower exposure to the chemical(s) with the greatest change in the permissible exposure limit (PEL). It was the judgment of the experts involved that by assigning controls for the "major" chemicals, exposures for all other chemicals would be controlled. Chemicals and/or processes not included in the proposed standard (e.g., those covered by separate 6(b) rulemaking) were excluded from the survey. Examples of chemicals not included in the survey are asbestos, formaldehyde and benzene.

Survey information collected from each respondent included:
* Type of processes at the establishment;

* Type and amount of chemical used, manufactured, or generated in each process;

* Number of work stations and workers related to the process;

* Potential chemical exposure above the proposed standards (monitoring data, recorded overexposures) at the process;

* Process location (indoors/outdoors), and configuration (size, full enclosure, partial enclosure);

* Ventilation or other controls in place; and
* Economic and other characteristics of the plant.

A computer algorithm was developed to assess survey data to determine if potential worker overexposure and therefore compliance costs occur for each process at an establishment. Figure G-2 presents a general diagram of the computer logic adopted for use in the survey. The logic assesses potential overexposures on the basis of: actual reported monitoring data; statements that overexposures occur; and the particular process location, configuration, type and amount of chemical use and existing controls in place.


    Figure G-2.  Computer Logic for Deriving Industry Cost of Compliance.

(For Figure G-2, Click Here)

(For Figure G-2 cont, Click Here)

When a respondent provided actual monitoring data for a process that indicated chemical exposures above the proposed standard, compliance costs were assigned to that process on the basis of prescribed controls for the given process. Where no monitoring data or reports of overexposure were available, the computer algorithm logic examined process and chemical characteristics to determine if workers at the process were potentially exposed to chemicals at levels over the proposed standard. The logic assessed the controls reported to be in place at the process and compared them with a list of controls thought necessary to control exposures in that process within the industry. When the required controls were reported to be in place, no compliance cost was assigned. When the required controls were not reported to be in place, a compliance cost per work station was assigned.

The computer algorithm determined that some processes within plants had no overexposures and consequently no compliance costs. Zero compliance costs resulted where no processes and/or chemicals were reported to occur at the establishment. Zero compliance costs also resulted when the respondent had monitored a process using ACGIH or NIOSH standards and found no overexposures. Where only very small quantities of chemicals were present in a process, none of which had a "major" proposed exposure limit changes, no overexposure was determined and zero compliance costs were assigned. The major/minor designation was based on the proposed change in the PEL (over or under a 50 percent decrease) as well as chemical characteristics such as form, particle size, and vapor pressure.

Process configurations and location also were indications of compliance. Processes which were reported as completely enclosed with no worker entry were assumed to be in compliance with the proposed standard (have no compliance cost). Outdoor loading/offloading processes or other outdoor/processes with no chemicals with "major" proposed exposure limit changes were assumed not to require control equipment and costs. Zero compliance costs were also assessed where processes which required control equipment reported that the prescribed equipment was currently in place.

An example of a process which was assigned a cost of compliance to install engineering controls is a coating and spraying process in SIC 2511, Wood Household Furniture. The survey respondent reported that toluene, n-butyl alcohol and xylene were used in this operation. The proposed standard for toluene reduces the existing PEL by 50 percent. This reduction is considered to require concerted exposure control and is considered a "major" proposed exposure limit change. Because workers were involved in the process and the process was reported to be neither located outdoors nor fully enclosed, controls were assumed to be necessary to insure compliance with the proposed standard. The control required for controlling exposures at this process was determined to be local ventilation. The type of local ventilation prescribed in this case is a spray booth at an estimated cost of $3,070 annually per work station. Because the respondent reported no local ventilation, the cost was assigned for the eight work stations reported, resulting in a total estimated annual cost of $24,560 for this process at this site.

Expert engineering and industrial hygiene judgment was used to determine which of the various controls would be necessary to control for exposures by process in the affected industries. Engineering controls identified included exhaust ventilation (local and general), process enclosure, and process change. Some or all of these will be required by affected plants for compliance with the proposed exposure levels. In addition, personal protective equipment such as respirators will be needed for intermittent maintenance activities where engineering controls are not feasible.

The engineers and industrial hygienists classified the approximately 180 specific processes identified in the survey into about 30 process groups for the purpose of assessing required controls and estimating costs. These process groupings were based on similarities in the processes and levels and types of exposures resulting from the processes. Factors used to group processes include the chemicals generally involved in the process, type and usual configuration of the equipment, usual workstation design, level and route of exposure, industry group where the process exists and worker tasks in relation to the equipment and exposure route. The process similarities translated into likenesses in required controls such as type of ventilation hood, booth or enclosure, air flow rates, duct configuration and type and size of filters or scrubbers. Organizations presenting process data to the docket that varied from that derived by OSHA are referenced in the specific industry descriptions in this chapter.

The compliance cost framework is presented in Table F-1. This table presents the process groups, the industries where the processes were identified, the general classification of controls specified and work station unit costs for the required controls assigned.


TABLE G-1: COMPLIANCE COST FRAMEWORK AND WORK STATION UNIT COSTS
_________________________________________________________________________
                                                              ANNUAL COST
                                INDUSTRY     REQUIRED CONTROL    PER WORK
PROCESS GROUP (1)           GROUP (SICs)    CONFIGURATION (2)     STATION
__________________________  ____________    _________________ ___________

Leather Processing, major             31    Local Ventilation    $  2,510
Leather Processing, minor             31  General Ventilation    $    720
Electrical & Electronics
  Manufacture                         36    Local Ventilation    $  2,520
Printing Processes, minor    27,38,73,80    Local Ventilation    $  1,240
Printing Processes, major          27,39    Local Ventilation    $  1,380
Glass Processing, major               32    Local Ventilation    $  3,890
Glass Processing, minor            32,36            Enclosure    $     90
Resource Recovery & Water    28,29,33,49        Enclosure and
  Treatment, major                          Local Ventilation    $ 21,900
Resource Recovery & Water             26        Enclosure and    $ 14,000
  Treatment, minor                          Local Ventilation
Foundry Operations, major             33    Local Ventilation    $  2,520
Foundry Operations, minor          33,39    Local Ventilation    $  1,820
Grinding, Blasting and       25,33,36,39    Local Ventilation    $  7,200
  Metalworking, major
Metalworking & Welding          All SICs    Local Ventilation    $  1,140
Coke Ovens                        29 (3)    Enclosures, Local    $150,000
                                            Ventilation & Air
                                                    Purifiers
Paper Manufacturing, major      26,30,39    Ventilation & Air    $  2,900
                                              Purification in
                                                Control Rooms
Paper Manufacturing, minor      All SICs    Local Ventilation    $    180
High Temperature Drying         All SICs    Local Ventilation    $  4,740
Layup                     3632,3715,3732    Local Ventilation    $ 16,550
                               3792,3995
Coating, Spraying, and          All SICs    Local Ventilation    $  3,070
  Adhesive Application
Chemical Handling and           All SICs    Local Ventilation    $  1,760
  Formulation
Material Handling and           All SICs  Local Ventilation &    $  1,120
  Inspection, major                         Partial Enclosure
Material Handling and           All SICs  General Ventilation    $    560
  Inspection, minor
Cleaning & General              All SICs    Local Ventilation    $  2,410
  Solvent Use, major
Cleaning & General              All SICs    Local Ventilation    $    710
  Solvent Use, minor
Waste Collection and           4953,5093      Respirators (4) $520/worker
  Transport
Painting, Maintenance           All SICs      Respirators (4) $520/worker
Welding, Maintenance            All SICs      Respirators (4) $520/worker
Sanding & Drilling/Boring          24,25    Local Ventilation    $  2,200
Cutting, Sawing & Planing          24,25    Local Ventilation    $  1,900
Zero Cost Processes:
    Laundering                        72
    Embalming                         72
    Permanents                        72
    Anesthesia                        80
_________________________________________________________________________
  Footnote(1): The "major" and "minor" designation of process groups
refers to the level of the exposure change and consequently the extent of
required control configuration costs within a given control and process
configuration.  For example, leather processing is the general process
group and processes within that group are classified based on whether the
employee exposure control requires major or minor control costs.
  Footnote(2): The specific required control configuration cost was
estimated including all necessary components, such as ductwork, fans,
hoods, baghouses, etc.
  Footnote(3): Coke ovens in SIC 33 are not included as they are covered
by OSHA's Cook Oven Standard.
  Footnote(4): Use of respirators is considered the only feasible control
for these processes due to their intermittent performance and because they
are generally not performed at a fixed site.

The development of unit costs for each control configuration required the development of "model" control designs. Model control configurations were selected to provide exposure control at "typical" process/work stations within the specified process group. This costing approach ("model" configurations for "typical" work stations) required the differentiation of some process groupings as major or minor. The major/minor differentiation addresses the expected level of control required.

The control designs were developed by engineers based on their experience in industry and extensive secondary research on operations and exposure situations in each industry sector. This research included an examination of industry and industrial hygiene journals, engineering process reports, and texts. Included in the detailed cost calculations for the control configurations were costs for enclosure construction, baffles, fans, ductwork, filters, scrubbers, baghouses, and all other equipment required for exposure control. All of the costs were developed on a per work station basis so that an average size did not need to be estimated for the process. Investment costs were assigned to each control design on the basis of engineering handbooks and supplier catalogs. Investment costs were annualized over the projected life of equipment (10 years) using a 10 percent cost of capital and adding annual operating and maintenance costs estimated at 10 percent of the capital cost. Respirator costs for use by maintenance workers for intermittent activities were considered annual costs and include the respirator purchase as well as an estimated year's worth of cartridges and canisters.

Process control costs were summed per establishment and any maintenance worker respirator costs were included. A total annualized capital cost and annual operating cost was developed for each establishment. Costs for the survey establishment were then weighted (by SIC and size) to represent compliance costs for the universe of affected plants.

The United Auto Workers (UAW) International Union contended that the OSHA analysis "establishes a far outer bound" for the costs of compliance for several reasons [Ex. 197]. Two reasons claimed by the UAW are that the survey failed to account for the current state of control of process units and that not all process units would require the full application of the control schemes specified in the OSHA analysis. These potential problems were explicitly considered in developing the estimation method and the method was designed to minimize the effects of these factors. Questions in the survey did ask about the controls in place for every process. But the mere presence of controls does not assure the ability to achieve proposed levels. OSHA believes that the assignment of full control costs to uncontrolled processes, although not always necessary, is approximately offset by not costing, in all situations, the upgrading of insufficient control systems already in place.

The UAW also contended that "OSHA has refused to collect readily available exposure data which would have supported the feasibility of much lower PELs." On the contrary, OSHA solicited exposure data in two ways, as well as searching for data from public agencies. In addition to asking for data in the public hearings and for submission to the docket, every survey respondent was asked to provide exposure data. In addition, data in OSHA's IMIS database, in NIOSH reports and in journal articles were used.

The fourth point made by the UAW to support its position that costs were overestimated is that "The ongoing replacement of plant and equipment has not been accounted for." OSHA did consider this factor where information was available which allowed a quantitative assessment of the effect it would have on compliance, such as in dry cleaning, although it could not be considered in all areas of industry.

Finally, the UAW disapproves of the method by which unit costs were estimated, claiming that OSHA's approach "degrades the value of the analysis by obstructing generalization. In addition, past cost estimates have been sensitive to a per cfm cost of ventilation. The present evaluation fails to present such a cost." Estimation of costs on a per cfm basis was avoided because OSHA felt that better estimates could be made by estimating more detailed unit costs. Rather than having only one per cfm cost, the OSHA analysis uses 30 control scheme costs which are able to take into account different characteristics of both the process equipment and the chemicals being controlled. OSHA believes that this method creates the ability to estimate costs much more accurately than a per cfm estimate would allow across the broad spectrum of industries and processes which this rule affects. The per cfm basis of cost estimation was not used because it would require much broader assumptions about average characteristics of control systems such as ductwork, baghouses, etc. OSHA views its method as an improvement on the previous methods because this method requires fewer generalizations and assumptions and allows the inclusion of more information in estimating costs.

Compliance Costs by Industry Sector

Following the methodology described in the preceding section of this chapter, annual compliance costs were estimated by industry sector. The costs presented for the surveyed industries are based on the data collected from the about 5,700 respondents. (For industries not included in the survey, expert judgment and secondary sources were used for estimating costs.) Table F-4 (shown at the end of the chapter) presents the detailed breakdown of compliance costs for each industry sector included in the survey. The table illustrates the processes reported in the survey, the number of work stations by process, and the number of work stations determined to require the addition of compliance controls. The process and work station frequencies are weighted to reflect the total universe of affected plants.

A small percentage of respondents (less than 5 percent) actually provided monitoring data during the survey. However, based on information from the survey, it was determined that about 86 percent of all establishments in the surveyed industries (74 percent of those with hazardous substances) have no exposures in excess of the final standard and will not incur any costs to comply with the standard. This conclusion was derived by comparing controls in place with controls deemed necessary to reduce exposures to the regulated limits. Thus a cost was assigned if the existing ventilation system was estimated to be insufficient to control these chemicals at the new levels. About 22 percent of establishments with hazardous substances will incur costs to provide engineering controls for processes within the plant. About 4 percent of the establishments with hazardous substances will be required to provide personal protective equipment only for maintenance workers whose intermittent operations cannot be controlled with engineering controls.

Table G-2 presents the total annualized capital and annual operating cost for compliance with the standard by industry. As shown, annual compliance costs are estimated to total $788 million. Upon review and incorporation of all docket materials, OSHA believes that these costs are fully representative of costs of compliance with the standard. These costs represent an estimate of compliance costs for large and small plants affected by the exposure limit changes. Industries with some anticipated cost impact are identified below. Included in the industry description are data provided to OSHA during testimony at the public hearing and in the docket submissions.


Table G-2: ANNUAL OPERATING AND ANNUALIZED CAPITAL COST OF COMPLIANCE BY
           INDUSTRIAL SECTOR (a)
_________________________________________________________________________
SIC (b)  SIC DESCRIPTION      LARGE PLANTS    SMALL PLANTS    ANNUAL COST
_______  __________________   ____________    ____________   ____________
20       FOOD PROD. (c)        $21,704,100     $11,789,000    $33,493,100
21       TOBACCO (c)               $19,700              $0        $19,700
22       TEXT. MILL (c)        $23,308.400      $6,170,000    $29,478,400
23       APPAREL PROD. (c)     $23,604,300      $8,139,900    $31,744,200
24       LUMBER & WOOD         $18,112,100     $38,608,700    $56,720,800
25       FURNITURE             $18,440,900      $2,634,900    $21,075,800
26       PAPER PROD.           $29,807,900      $1,190,800    $30,998,700
27       PRINTING & PUB.        $5,186,400     $28,568,100    $33,754,500
28       CHEMICAL PROD.        $28,793,500      $6,661,200    $35,454,700
29       PETRO. REFINING       $23,635,000         $51,000    $23,686,000
30       RUBBER & PLASTICS     $46,605,200     $64,488,200   $111,093,400
31       LEATHER PROD.          $1,272,000      $1,142,000     $2,414,700
32       STONE & CLAY          $15,704,300      $6,753,500    $22,457,800
33       PRIM. METAL           $65,691,400      $5,266,200    $70,957,600
34       FAB. METALS           $28,964,500     $10,455,200    $39,419,700
35       MACHINERY             $30,994,600     $14,212,000    $45,206,600
36       ELEC. MACH.           $17,060,100      $3,607,400    $20,667,500
37       TRANS. EQUIP.         $23,577,900     $26,214,500(c) $49,792,400
38       INSTRUMENTS            $7,227,700      $2,405,800     $9,633,500
39       MISC. MANUF.           $9,829,300      $6,013,300    $15,842,600
40       R.R. TRANS.            $1,083,400              $0     $1,083,400
45       AIR TRANS.             $3,740,500              $0     $3,740,500
47       TRANS. SERV.           $3,789,400              $0     $3,789,400
49       ELEC. GAS. SAN.       $32,355,100      $5,654,200    $38,009,300
50       WHOLESALE TRADE        $1,306,700      $1,688,600     $2,995,300
51       WHOLESALE, NON-DUR     $2,854,900     $11,360,900    $14,215,800
55       AUTO DEALERS           $7,382,200      $6,168,300    $13,550,500
72       PERSONAL SRV.          $3,487,100      $7,385,000    $10,872,100
73       BUSINESS SRV.          $1,596,100        $826,000     $2,422,100
75       AUTO REPAIR            $4,280,100      $1,863,400     $6,143,500
76       MISC. REPAIR SRV.        $469,700      $2,340,200     $2,809,900
80       HEALTH SERV. (c)       $2,413,000      $2,026,400     $4,439,400
                              ------------    ------------   ------------
TOTAL                         $504,298,200    $283,684,700   $787,982,900
_________________________________________________________________________
Source:  U.S. Department of Labor, Occupational Safety and Health
         Administration, Office of Regulatory Analysis.

  Footnote(a):  Costs were calculated by annualizing the capital cost over
the projected life of the equipment (10 years) using a 10 percent cost of
capital and adding an annual operating and maintenance cost estimated at
10 percent of the capital cost.
  Footnote(b):  Industry sectors not identified in this table include
industries with no major cost impact expected, the construction industry,
which will be the subject of a separate regulatory analysis, and
industries such as mining, over which OSHA has no jurisdiction.
  Footnote(c):  Costs in these sectors were based on expert judgement and
secondary data collection.

In addition to review of submitted materials, OSHA also undertook a site visit survey of about 90 plants to examine and compare data collected by telephone and in the field. Various statistical tests were performed on the telephone survey and site visit data to detect biases in the cost algorithm [Supplement 1]. These analyses tested the hypothesis that the telephone survey did not systematically differ from the site visits in the number of estimated firms out of compliance and the actual cost assigned to those firms. Using a 95 percent confidence interval, these tests revealed no aggregate bias in the assignment of costs by the telephone survey as compared to the site visits.

Food and Kindred Products (SIC 20). Costs are projected for a large number of establishments in this sector. The prepared feeds and feed ingredients, not elsewhere classified (SIC 2048), are estimated to account for a large percentage of the $33.5 million annual costs in SIC 20. Controls may be necessary for dust exposures and chemical fumigants.

Two commenters provided detailed cost estimates for firms to achieve the Agency's proposed grain dust limit of 4 mg/m(3): the National Grain and Feed Association (NGFA) [Ex. 3-752] and the National Feed Industry Association (NFIA) [Tr. 8/10/88, pp. 10-61 - 10-69].

The National Grain and Feed Association provided alternative cost estimates for feed mills and for flour mills. These estimates were based on the following assumptions:

(1) All affected facilities will need pneumatic dust control systems and do not now have them;

(2) Only 13 percent of feed mills handle wheat, oats, or barley, and thus only 13 percent will be affected by the grain dust limits; and

(3) The costs of pneumatic dust control systems are the same as those estimated by Booz Allen in a 1984 study conducted for OSHA in connection with the Agency's grain handling standard (these costs were inflated by 15 percent to convert them from 1984 to 1988 dollars).

Using these assumptions, the NGFA concluded that the capital costs of compliance for all feed mills would be $213 million and for all flour mills would be $81 million [Ex. 3-752]. If these costs are annualized using OSHA's interest-rate and life-of-equipment assumptions, annualized costs for feed mills (using a 10-percent operating cost figure) would be $56 million per year, and annualized costs for flour mills would be $21 million per year. Average annualized costs per affected feed mill would be $44,000 per year and per affected flour mill, $225,000 per year.

The NFIA also provided estimates of the total costs of compliance for feed mills. The NFIA's costs were based on an evaluation of 20 existing feed mills; the NFIA, therefore, only attributed costs for pneumatic dust control systems to facilities that do not now have them. Thus, the NFIA study attempts to take into account baseline controls; it reported that 5 of these 20 mills had some level of dust control in place. The NFIA estimated that the capital costs of compliance for all feed mills would be $664.5 million [Tr. 8/10/88, pp. 10-61 - 10-69]. If OSHA's interest rate and life-of-equipment assumptions are used and operating costs are assumed to be 10 percent of capital costs, annualized costs for all feed mills would be $175 million per year. Average annualized costs per affected feed mill would be $22,000 per year.

OSHA's Preliminary Regulatory Impact Analysis (PRIA), published with the proposed rule, estimated costs for all facilities in SIC 20 to comply with the proposed PELs; the grain dust portion of this overall cost was based on the assumption that affected facilities would have to meet a 4-mg/m(3) PEL for grain dust. Based on health effects and economic feasibility considerations, the Agency has changed the grain dust levels to 10 mg/m(3) TWA. OSHA's analysis shows that the great majority of employee exposures are at or below 10 mg/m(3), and thus, few additional controls will be needed to achieve the final rule's limit of 10 mg/m(3). OSHA believes that the cost estimates for this sector are conservative and probably overstate the costs that affected employers will be required to expend to achieve the 10 mg/m(3) limit.

Several comments were received from employers in SIC 201, Meat Products, who were concerned that the proposed limit of 1 ppm for carbon disulfide would force the manufacturers of the cellulosic casings that are made in facilities classified in SIC 3089, Miscellaneous Plastics, to go out of business [Exs. 3-421, 3-659, 3-897; Tr. 8/2/88, pp. 4-209, 4-261]. In the opinion of these concerned meat packers and processors, the impact of the carbon disulfide limit on firms in SIC 3089 would be so great that all domestic supplies of the cellulosic casings needed by the meat packers and processors would disappear. These commenters were particularly concerned because there are no substitutes for cellulosic food casings except natural casings, which can only be used for cooked sausage and cannot be used with automated machinery [Ex. 3-897].

The costs anticipated by the meat packers and processors were presented in a study conducted by Wharton Economic Forecasting Associates (WEFA) [Ex. 3-659]. WEFA forecast a loss of 12,000 to 20,000 jobs in meat processing and 12,000 to 16,000 jobs in meat packing, and also projected an 8 and 16 percent reduction in the price paid to farmers for cattle and hogs, respectively [Ex. 3-659].

WEFA based its forecast on the following assumption: that processed meats dependent on cellulosic casings would disappear from the marketplace altogether [Ex. 3-659]. OSHA finds this scenario unlikely, since the cost impacts in SIC 3089 are likely to be greatly reduced because the Agency has established a 4 ppm limit, rather than the proposed 1 ppm limit, for carbon disulfide in the final rule. Domestic production of casings should not cease or be disrupted in a major way.

OSHA concludes that the concerns of the meat packers and processors in SIC 201 have been addressed and their supplies of cellulosic food casings should not be disrupted.

The National Cotton Council of America [Ex. 3-1080] expressed the concerns of its members over the difficulty small, rural cottonseed mills would have in sampling their employees' exposures to hexane and grain dust. As discussed above in the section on Technological Feasibility for SIC 20, OSHA determined that sampling and analytical methods are available for these contaminants and that consultant industrial hygienists can be employed by mill owners on an as-needed basis. OSHA is aware that the services of competent and experienced industrial hygienists can be obtained for fees beginning at $300 per day and that laboratory fees for analysis range from $20 to $40 per sample, depending on the substance being analyzed. OSHA does not believe that costs of this magnitude will have a significant impact on cottonseed mills.

Although carbon dioxide exposures in the beer industry were described as "unique" [Tr. 8/9/88], the principal sources of exposure are blow-outs of safety valves, opening of tank doors, and entry into tanks for cleaning. For both blow-outs (an upset condition) and tank entry (a maintenance operation), OSHA permits the use of respiratory protection to meet the PEL. Exposures resulting from opening tank doors can be reduced by implementing the work practice of cracking the door and remaining out of the area for a few minutes to allow the CO(2) to dissipate. Since the final rule establishes an 8-hour TWA of 10,000 ppm, rather than the 5,000 ppm proposed, OSHA concludes that the cost estimates presented in the PRIA for SIC 20 do not need to be revised and include all potential costs of compliance for breweries.

The Corn Refiners Association (CRA) estimates that the wet corn milling industry would incur $24,097,000 in capital costs, with annual operating costs of $6,244,000 million, to meet the proposed 2- and 5-ppm SO(2) standard. Of this, CRA estimates that $12,809,000 in capital costs and $3,266,000 in operating costs would be incurred to meet the 5-ppm STEL, and $11,288,000 in capital costs and $2,878,000 in operating costs would be incurred to meet the 2-ppm PEL [Ex. 65, Tab 13, pp. 7-8].

OSHA notes that 47 percent of CRA's estimated costs are attributed to meeting a STEL. However, the record indicates that short-term excursions do not typically occur during normal operations; instead, they occur during maintenance activities and in emergency conditions. In these situations, the standard practice in the industry is to use respiratory protection [Tr. 8/8/88, p. 8-90], as would be permitted by OSHA.

Furthermore, OSHA's technological feasibility assessment shows that the 2-ppm TWA and 5-ppm STEL can be achieved in all routine operations in this sector with the addition of a small amount of make-up air (or by opening the windows in warmer months). Employers in this sector also need to reduce the number of process upsets and maintenance problems in their plants by instituting manual leak detection programs, improving maintenance, replacing pump seals before they leak, and phasing out outdated process equipment. Evidence in the record reports that the volume of production and sales has risen so quickly that control equipment has been unable to keep pace [Ex. 65, Tab 13, p. 7]; this sector should therefore not have difficulty absorbing the negligible costs associated with the minimal control procedures needed for this sector to achieve compliance with the final rule's limits for SO(2).

CRA notes that its estimates of costs constitutes 18.3 percent of OSHA's total cost estimate for all chemicals in all parts of SIC 20. OSHA notes that, generally, within a 2-digit SIC industry group, most industry sectors are estimated to incur minimal costs to comply with the final rule, and a few industry sectors will incur higher costs. Thus, even assuming that CRA's estimated costs are accurate, OSHA's aggregate estimate for SIC 20 are not necessarily substantially understated. Thus, OSHA concludes that the cost estimates presented in the PRIA for SIC 20 do not need to be revised based on the record evidence pertaining to the potential costs of compliance for wet corn milling.

Tobacco Manufactures (SIC 21). The lowest cost of compliance in the manufacturing sector is expected to occur in SIC 21, Tobacco Manufacturers, ($20,000). It is estimated that very few plants will incur costs in the tobacco manufacturing industry.

Textile Mill Products (SIC 22) and Apparel and Other Finished Products (SIC 23). These sectors have a large number of establishments which may incur compliance costs. The apparel industry is estimated to incur about $31.7 million in annual compliance costs. Many of the affected establishments in SIC 23 may require controls for cleaning solvents such as perchloroethylene. The $29.5 million annual costs in the textile industry are estimated to result from control of exposures to solvents, dyes and other substances. No differing cost estimates in opposition to OSHA's cost calculations were presented in the docket or testimony.

Lumber and Wood Products (SIC 24). The annual costs of compliance in the lumber and wood products industry are estimated to total $56.7 million. The compliance costs for this sector primarily reflect the cost of controls required to lower exposures of wood dust to 5 mg/m(3) (2.5 mg/m(3) for Western red cedar wood). The survey indicated that sanding and other "dusty" processes would require controls to lower wood dust exposure. The large number of establishments that must engineer ventilation systems for wood dust control account for the substantial proportion of compliance costs to be incurred by small establishments in this sector.

OSHA's estimates in the preliminary analysis were based on a standard of 5 mg/m(3) for softwood and 1 mg/m(3) for hardwood, using survey responses for particulates not otherwise regulated as a surrogate for wood dust.

In determining the total cost of compliance for wood dust at the final PEL of 5 mg/m(3) (2.5 mg/m(3) for Western red cedar), OSHA carefully considered data presented in to the record by National Economic Research Associates (NERA), Clayton Environmental Consultants, the Workers' Institute for Safety and Health (WISH), the Holliday report, and numerous other government, union, and industry respondents [Exs. 3-748, 8-127, 8-196, Tr. 8/18/88, p. 13-5, etc.]. Researchers from Clayton Environmental Consultants and NERA, on behalf of the Inter-Industry Wood Dust Coordinating Committee, performed a study on the impacts of the proposed air contaminants rule on SICs 24 and 25 [Exs. 3-748, 8-127]. NERA concluded that it would cost firms in SICs 24 and 25 $266 million annually for a 5 mg/m(3) standard for all woods. Under a 1 mg/m(3) standard for all wood dust, NERA estimated that costs would exceed $1.9 billion annually and under the proposed standard of 5 mg/m(3) for softwood and 1 mg/m(3) for hardwood, annual compliance costs would be approximately $1.5 billion. NERA's estimate of $1.5 billion was more than four times higher than OSHA's August 1st estimate of $341 million annually for a 5 mg/m(3) softwood standard, 1 mg/m(3) hardwood standard [Ex. 38a]. Mark Berkman, representing NERA, testified that the cost discrepancies between the OSHA study and their estimates were due to the differences in unit costs and in the number of work stations out of compliance in SICs 24 and 25 [Tr. 8-12-88, p. 107, 111].

OSHA determined that annual unit costs of compliance per work station of $1,900 for cutting/sawing/planing, $2,200 for sanding/polishing and grinding, and $2,200 for drilling/boring are the best estimates currently available to comply with the final standard. These unit costs are not significantly different from the unit costs presented in the Clayton study. (The unit cost presented by NERA in one case does not accurately reflect the findings of Clayton Environmental Consultants. Apparently the cost applied to the "belt sander" in the NERA study was derived for "sander, belt (widebelt)" in the Clayton study. However, the cost developed for "sander, edge" would have been more appropriate. "Edge sander" was never identified in the NERA survey. The cost for the widebelt sander is $50,800, while that for the edge sander is $12,900. The capital cost for control on a belt sander developed for OSHA was $8,000.) In analyzing NERA's methodology, there were significant differences between OSHA's estimates of costs and work stations when compared to NERA's. NERA's methodology begins by surveying "industry experts" (via the Inter-Industry Wood Dust Coordinating Committee) to derive the number of machines in typical small and large establishments. In a number of industries, these experts estimated that there would be many more machines than total employees. For example in SIC 2426, NERA's survey respondents estimated that there would be 32 machines in a "typical" small plant (fewer than 20 total employees). NERA's next step was to multiply the number of machines in a typical plant by the percentage of machines out of compliance from the Clayton study. This revealed an estimated number of machines out of compliance in a typical plant at the four-digit level for small and large firms. This estimate of machines out of compliance was then multiplied by the per machine unit cost to arrive at an average cost per typical plant. Finally, this number was multiplied by the number of plants reported in the 1982 Census of Manufactures for each four-digit SIC in order to arrive at an aggregate cost.

OSHA believes that its methodology for deriving total work stations is more accurate. OSHA used a telephone survey which requested information about work stations specifically at the plants being interviewed. However, NERA sent surveys to "industry experts" who were asked to describe a "typical plant". NERA never explains the number or identity of respondents in its survey. However, it is important to note that NERA received no responses in numerous four-digit SICs (large plants in SICs 2429, 2491, 2515, 2517, 2519, 2531, and 2541 and small plants in SICs 2436, 2451, 2452, 2491, 2512, 2515, 2517, 2531, and 2541). For industries with no response rate, a weighted average cost was used as a surrogate. In the case of SIC 25, 10 of the 16 size and industry categories were derived by using surrogates. Thus non-surveyed industries such as mattresses and bedsprings were estimated to have the same costs and work stations out of compliance as surveyed industries with high wood-dust-generating processes such as wood and upholstered furniture. The cost surrogate used for these 10 categories in SIC 25 is the third highest per plant cost, despite the fact that it was derived without specific exposure data by Clayton or estimates of machines used in a typical plant for these specific four-digit SICs. OSHA concludes that such extrapolation is based on less comprehensive data than the 1988 telephone survey. This widespread use of surrogates partly explains why total work stations has been overestimated by NERA.

Additionally, OSHA concludes that NERA's survey estimates of total machines is high, and therefore the number of machines out of compliance is overestimated. Including those four-digit SICs where surrogates are used (and therefore total estimated number of machines is implied), NERA assumed a total of approximately 800,000 machines in SICs 24 and 25. This estimate is roughly equivalent to the number of employees in these two SICs. The statement by NERA that "... workers (or work stations, assuming one worker per work station)" [Ex. 3-748, p. 13] implies that NERA does not find it unreasonable that 800,000 wood-dust-generating machines are used continually by every worker in SICs 24 and 25. OSHA concludes that its estimate of 300,000 total work stations (200,000 wood-dust-generating work stations) as derived from the 1988 telephone survey, is a more accurate estimate. OSHA's site visits and survey indicate that there are far fewer work stations than workers in SICs 24 and 25. One cause for this difference is the amount of shift work performed, thus allowing one work station to be used by two or three workers in a single day. Another cause for this difference is the number of technical, clerical, managerial, and maintenance staff, many of whom do not work consistently around machines which generate substances regulated under this rulemaking. Thus OSHA has determined that its estimate of total work stations is an accurate assessment for firms in SICs 24 and 25.

Next, it was necessary for OSHA to derive the percentage of wood-dust-generating processes (sanding/polishing/grinding,cutting/sawing planing, and drilling/boring) out of compliance with the final standard in SICs 24, 25, and 26. OSHA combined its monitoring data from site visits with Clayton's samples to estimate that 16 percent of the wood-dust-generating work stations (including those involving Western red cedar) would be out of compliance with the final standard. This percentage seems to be reasonably close to the 13.5 percent figure for 5 mg/m(3) from the OSHA Health Response Team Survey referenced by Scott Schneider of the Workers' Institute for Safety Health (WISH) [Tr. 8/15/88, p. 13-5].

Since a 2.5 mg/m(3) standard was established for Western red cedar, OSHA performed a separate analysis on compliance cost for this substance. OSHA believes that there are approximately 290 firms involved in the production of shakes and shingles with Western red cedar in SIC 2429 [U.S. Department of Commerce, Office of the Census]. Studies on Western Red Cedar asthma [Ex. 82D, Captain James J. Edwards, Jr.] indicate that approximately 90 percent of these firms operate in Washington State, where the permissible exposure limit is currently 2.5 mg/m(3). Data presented by Stephen Cant of the Washington Department of Labor & Industry indicated that "they can, in fact, in most cases, comply with those limits, and that there are studies that support, certainly I think, the 2.5 limit as regards allergenic wood dust with respect to Western red cedar." [Tr. 7/29/88, p. 2-103]. However, studies performed by the University of Washington in 1987 indicate that "Labor and industries inspectors found a large number of mills out of compliance with the new regulatory standards." [Ex. 127.H] OSHA assumed that compliance with the wood dust standard relative to Western red cedar in the shakes and shingles industry would not be significantly different from compliance with the overall wood dust standard. OSHA concluded that 16 percent of the work stations would be out of compliance with the final standard in the shakes and shingles industry.

To derive the cost for wood dust in SIC 24, OSHA estimated that 142,000 of the 215,000 total work stations are wood-dust-generating, and that 1500 involve Western red cedar. Sixteen percent, or 23,000, of the wood-dust-generating work stations were determined to exceed the final standard (240 for Western red cedar). Wood dust thus accounted for $45 million of the $56 million in SIC 24. For reasons explained above concerning the total number of work stations and work stations affected, OSHA concludes that NERA's estimate of $137.1 million for a 5 mg/m(3) standard is an overestimate.

In addition to wood dust, controls for exposures to solvents, wood preservatives, and other chemicals in coating processes are estimated to result in compliance costs in SIC 24. Overall, about 68 percent of all establishments in SIC 24 are estimated to incur compliance costs. OSHA thus concluded that the annual operating and annualized capital cost to comply with all standards would be $56.6 million in SIC 24.

Furniture and Fixtures (SIC 25). Annual costs of compliance in the furniture and fixtures industry are estimated to total $21.1 million. Costs to control wood dust exposures at 5 mg/m(3) wood (2.5 mg/m(3) for Western red cedar) during sanding, cutting, drilling, and other dusty processes are the major components of compliance costs in this sector. Establishments would also incur costs for control of exposures to coatings and solvents. The survey indicated that the furniture sectors which include metal working (SICs 2514, 2515, 2522, 2542, 2591 and 2599) would also require controls for welding fumes and various metal particulates resulting from grinding and other processes. OSHA believes that local exhaust ventilation will reduce exposures to permissible levels during welding operations.

OSHA again believes that NERA overestimated the costs and the number of work stations used in its cost estimations for SIC 25. An explanation of cost differences is provided in SIC 24. OSHA calculated 89,000 total work stations (57,000 at wood-dust-generating work stations) in the furniture industry, based on responses provided by the telephone survey. NERA's estimate of work stations, which relied heavily on surrogates, resulted in a significant overestimate of the number of total work stations and work stations out of compliance with the OSHA standard. This overestimation of the total number of work stations distorted NERA's cost estimates for SIC 25 ($128.9 million for a 5 mg/m(3) standard).

To derive the cost to control wood dust exposures in SIC 25, OSHA estimated from the 1988 telephone survey that 57,000 of the 89,000 total work stations would be wood-dust-generating. Sixteen percent, or 9,000, of the wood-dust-generating work stations were expected to exceed the final standard. Wood dust thus accounted for $19 million of the compliance cost in SIC 25. OSHA believes that its total cost estimate of $21.1 million is an accurate estimate of the actual cost of compliance for this sector.

Paper and Allied Products (SIC 26). Annual costs in the paper and allied products industry are estimated to be $31.0 million. Much of the estimated costs in SIC 26 will be associated with the cost of controls in large pulp mills and associated operations. Pulp mills are operated separately (those listed in SIC 2611) or as part of paper or paperboard mills (SIC 2621 and SIC 2631 respectively). Some of the cost of compliance in these operations would result from controlling the large quantities of chemicals used in breaking down the pulp to form cellulose and the reactions that occur in the digesting process. The digesting and bleaching operations require ventilation or enclosure.

A portion of the costs associated with SIC 26 relate to controlling exposures to wood dust levels at 5 mg/m(3) for wood dust (2.5 mg/m(3) for Western red cedar). Data presented on wood dust exposures by Clayton were derived from only 2 site visits in SIC 26. NERA presented no cost estimates for this industry. Thus OSHA retained its estimate that sixteen percent of all wood-dust-generating work stations would be out of compliance with the final standard in SIC 26. Data from the 1988 sample survey indicated that the total cost for this SIC would amount to an annual operating and annualized capital cost of compliance of approximately $31.0 million.

Printing and Allied Industries (SIC 27). Compliance costs in the printing industry sectors (an estimated $33.8 million) would result from ventilation requirements to control exposures to cleaning solvents and ink spray generated within the printing process. A very large number of small establishments are involved in printing and over 3,100 of them would be affected by the revised standards. The survey indicated that a large number of small establishments currently lack exposure controls and provision of these controls accounts for the high control costs in this sector. However, OSHA's field visits in this sector [Ex. 8-11] indicated that the unit costs initially estimated for printing processes were somewhat high. The final cost estimate was adjusted to reflect the information collected during the field site visits.

Chemicals and Allied Products (SIC 28). Annual compliance costs in SIC 28 are estimated to total $35.5 million. Over 35 percent of the costs in SIC 28 are estimated to occur in Paints and Allied Products Manufacturing (SIC 2851). The survey indicated that a large proportion of plants will require additional controls for a number of processes found in paint and paint products manufacturing. There are many chemicals in this industry segment which present exposure problems in a variety of both wet and dry processes, including reaction, separation, crushing, mixing, drying and bagging.

According to U.S. Borax, the average annual operating costs for environmental controls at Borax in SIC 2819, Industrial Inorganic Chemicals (NEC), is considerably higher than OSHA had predicted for a large plant. As an example, OSHA estimated operating costs of $18,000 per year for large plants in SIC 28. U.S. Borax estimated an average operating cost of $37,600 per year. [Tr. 8/9/88, 9-113.] It is not clear from the testimony or from submissions to the docket [Ex. 3-744] which of the costs listed by Borax are associated with the mining and initial processing of the ore. These processes fall under the jurisdiction of the Mine Safety and Health Administration. OSHA believes a significant portion of these costs estimated by Borax are associated with the mining operation rather than downstream activities. After reviewing the rulemaking record, OSHA increased the TWA for all borates to 10 mg/m(3) and adjusted plant costs downward to reflect the change.

Industry group SIC 282, Plastics Materials, Synthetic Resins and Synthetic Rubber accounts for about 22 percent of compliance costs in this sector. Compliance costs are related to ventilation and other requirements to control exposures to carbon disulfide, acetone and other emissions in the manufacture of rayon, cellulose acetate fibers and other plastics materials and synthetic rubber. The Vinyl Institute contended that a number of the processes found in member companies would have to be modified at an estimated capital expense of $10-25 million. The additional annual expense to maintain the required level of compliance was estimated by the Institute to be $4-5 million. The industry-wide estimated initial capital expense was estimated to be $160-400 million and annual expenses $60-80 million [Ex. 3-624]. The cost estimates submitted by the Vinyl Institute included tank farm vent controls which OSHA, as explained in Chapter F (Technological Feasibility), concludes would not be necessary. The remaining areas identified by the Institute are loading/unloading operations and process sewer systems. While the costs are not presented in a disaggregated form, OSHA believes that the costs to bring these two areas into compliance would be only a fraction of the Institute's total cost estimate and OSHA's estimated costs are more accurate. OSHA also notes that the final limits for several chemicals of interest to the Vinyl Institute (acetone, carbon disulfide) are less stringent than those proposed, which should mitigate cost problems for affected firms.

In SIC 2823, Cellulosic Manmade Fibers, the Inter-Industry Committee on Carbon Disulfide asserts that "the cost of making even small improvements below the 20 ppm limit is significant -- $16.6 million." These costs would be for preventing the escape of carbon disulfide into the work area (process enclosures) and for increased ventilation [Ex. 3-747, p. 82]. As explained in the discussion of technological feasibility, OSHA believes that the evidence indicates the problem to be much less severe than is suggested above, and that most exposures are of short duration. The industry can comply with the final carbon disulfide standard of 4 ppm by using respirators in a limited number of designated processes (see Chapter F, Technological Feasibility) and adjusting work practices to control exposures. Costs for this subsector are reflected in the total estimate for SIC 28. Also in SIC 2823, the manufacturers of cellulose acetate claim that compliance with the acetone standard is not economically feasible at the three existing facilities. Tennessee Eastman estimated that costs of compliance with the proposed standard of 250 ppm for acetone to be $11.2 million annually in its facility [Ex. 3-745]. Mr. Vernon G. Knight of Hoechst Celanese, estimated that the costs of compliance for its two facilities would total $40.2 million in capital costs [Ex. 3-745]. OSHA does not believe that costs of this magnitude will be incurred. OSHA has revised its original proposal of 250 ppm for acetone. OSHA believes that a 750 ppm TWA and 1000 ppm STEL is economically and technologically feasible, and the costs for this sector have been reduced to reflect this change.

For SIC 2892, Explosives, the Institute of Makers of Explosives mentions "a study conducted in one nitroglycerin/ethylene glycol dinitrate (NG/EGDN) manufacturing facility in which the concept of reducing workplace concentrations to a 0.01 ppm (0.1 mg/m(3)) level was examined." This study indicated that the costs of engineering controls at this facility would exceed $4 million (1979 dollars) in capital costs to achieve the proposed standard for NG/EGDN [Ex. 3-749, 190]. OSHA believes that the principal cost for SIC 2892 would be for air line respirators and this cost is included in the total cost estimated for SIC 28.

Petroleum Refining and Related Products (SIC 29). Although only 13 percent of all facilities in SIC 29 are expected to be affected, nearly 25 percent of the large refineries will incur costs. Of those firms with more than 100 employees, almost 59 percent incurred some cost. Approximately 90 percent of the $23.7 million annual costs in SIC 29 are expected to be incurred by facilities in SIC 2911, Petroleum Refining. Most of these costs will be related to water treatment processes and sampling/quality control tasks because of a lack of controls in place in these two areas. In general, however, this industry has extensive control technology in place for the primary processing equipment. Closed processes with few exposed workers are predominant due to the requirements of process operation at elevated temperatures and pressures.

Costs in SIC 2951, Paving and Roofing Materials, arise mainly from smaller blending and formulating operations which usually involve few employees. Packaging and loading/offloading processes account for the majority of costs in SIC 299, Miscellaneous Products of Petroleum and Coal. The remainder of costs in SIC 299 are attributable to the blending and formulating of lubricating oils and greases.

Rubber and Miscellaneous Plastics Products (SIC 30). Annual costs of compliance in this industry sector are estimated to total about $111.1 million. Controls were required for processes such as molding and vulcanizing. Worker exposure to chemical vapors require the addition of local ventilation to many processes. The miscellaneous plastic products industry (SIC 3079) accounts for over 20 percent of the annual costs in this sector. The costs in SIC 3079 result from the high proportion of small plants in this sector which will incur costs of compliance. Controls are required in SIC 3079 for many crushing and grinding operations used to prepare plastic material for hot processes.

The Styrene Information and Research Council (SIRC) presented estimates of the costs SIRC believes will be required to control styrene exposures to a TWA of 50 ppm in selected segments of the miscellaneous plastics industry [Exs. 3-742, 34A, Tr. 8/3/88, pp. 117-130]. These cost estimates were developed for SIRC by Arthur D. Little, Inc. (ADL) and represent a partial update of a large study done by ADL in 1980 on the costs and technical feasibility of styrene control. This updated study estimated costs for the tub/shower, lavatory, hot tub/spa, and resin-applied-at-press segments. All of these segments are classified in SIC 308, Miscellaneous Plastic Manufacturing. ADL estimated that total capital costs for these segments to comply with an 8-hour TWA of 50 ppm for styrene would be $1.169 billion and that operating costs would be $204.6 million per year [Ex. 34A, Table 2]. Using OSHA's interest rate and life-of-equipment assumptions, the annualized costs for this sector, using these capital and operating costs, would be $395 million per year, a value in excess of the PRIA's total estimated costs of $75 million per year for all of SIC 30. ADL estimated that 19,230 employees in these segments are exposed to styrene at levels above 50 ppm; according to ADL, there are a total of 48,885 employees in these segments at 1550 plants [Ex. 34A, Table 1].

OSHA finds several difficulties with the ADL study. First, ADL used the exposure data from its 1980 study for SIRC as the basis for estimating what controls (and therefore costs) would be involved in achieving compliance; these exposure data showed considerably higher exposure levels (with one exception) than more recent data, e.g., the Cal/OSHA study of styrene exposures in this industry. ADL did use the Cal/OSHA data in one case (for the tub/shower segment), the only instance in which the Cal/OSHA exposure data were actually higher than the 1980 ADL data. Thus, ADL relied on the highest exposure data as a cost baseline, even when more recent data were available, and only used recent data when they were higher than the outdated data. This factor would contribute significantly to an overestimate of costs, especially since representatives of SIRC reported at the hearing that conditions in the industry have improved considerably in the last 10 years [Tr. 8/4/88, p. 5-94].

Second, compared with other sources (SIRC's prehearing submittal, the Cal/OSHA study), ADL estimates that many more workers are overexposed to styrene in these reinforced plastics industry segments. For example, ADL [Table 10, Ex. 34A] estimates that 20 percent of the workforce in the reinforced plastics segments of concern is overexposed to styrene, while the Cal/OSHA study [Ex. 3-742, Attach. 2, p. 30] reports that only 22 percent of the gel coat/lamination workers (who constitute approximately 36 percent of this work force) are overexposed. Thus, ADL used an inflated estimate of the number of overexposed workers in these segments; this factor also contributes substantially to an overestimation of costs.

Third, ADL seriously underestimates the existing level of baseline control in these segments. For example, ADL assumes that facilities have no control in place. However, as the Cal/OSHA study, SIRC testimony, and OSHA's site visits show, this is not the case. At the time of ADL's 1980 study, spray booths may have been nonexistent in these facilities, but that is clearly not the case today.

Fourth, ADL underestimates the effectiveness of the methods available to control exposure. For example, the Cal/OSHA study found that many facilities with only minimal levels of control were routinely achieving the 50-ppm limit and that others that were exceeding 50 ppm could achieve compliance by adopting minor engineering improvements, implementing better/maintenance procedures, and instituting improved work practices [Ex. 3-742, Attach. 2, pp. 20, 29-33]. Dr. Daniel Boyd, speaking for SIRC, testified to the effectiveness of improved work practices (training workers to leave the spray booth when not involved in sprayup/layup operations, to position themselves properly during spray operations, etc.). Based on his experience, Dr. Boyd estimated that work practices alone could reduce employees' 8-hour exposures to styrene by 50 percent [Tr. 8/2/88, p. 5-106].

Fifth, ADL ignored control approaches based on substitution, preferring instead to estimate that major revamping of ventilation systems and installation of local exhaust ventilation would be necessary in all facilities. OSHA is aware that other materials cannot be substituted for styrene in all applications; however, a costing methodology that relies exclusively on engineering controls ignores the movement in this sector away from styrene and high-emitting resins. A series of methodological problems, which compound each other, seriously undermines the usefulness of the ADL study. The Agency believes it more appropriate to rely on OSHA's industrywide survey as a source of data and to use the cost algorithm as a method of evaluating costs in these sectors. OSHA therefore concludes that the costs reported in the PRIA for SIC 30 are representative and reliable estimates.

The limit of 4 ppm for carbon disulfide may not be achievable with engineering controls in some operations performed during the manufacture of cellulosic food casings (SIC 308). These operations include unloading xanthate from the baratte, aligning of casing strands in the extrusion cabinet, and puncturing casings at the extrusion nozzle. Air-supplied hoods are currently used by workers performing these operations, and OSHA finds that respirators are likely to continue to be needed in these three processes, which require the opening of process machinery. Because employers will be permitted to use respirators to achieve compliance during these three operations, OSHA concludes that the cost estimates presented in the PRIA for SIC 30 accurately reflect costs for this industry.

Leather and Leather Products (SIC 31). One of the lowest costs of compliance in the manufacturing sectors is expected to occur in SIC 31, Leather and Leather Products ($2.4 million). In the leather and leather products industry sector, most of the affected establishments produce manufactured leather goods. The costs in this SIC are predominantly derived from gluing operations.

Stone, Clay, Glass and Concrete Product Manufacturing (SIC 32). The stone, clay, glass and concrete product industry is estimated to incur compliance costs of about $22.5 million. A major part of the annual costs in this industry segment may occur in the concrete, gypsum and plaster products (SIC 327) industries. According to the survey, controls in this sector are primarily expected to control silica generated during large scale crushing, grinding and sizing operations.

Primary Metal Manufacturing (SIC 33). The annual costs of compliance in primary metal manufacturing are estimated to total $71.0 million. The costs of compliance for this sector are heavily weighted by the cost of controls required in large establishments in this segment. Blast furnace establishments and primary foundries have large numbers of hot processes which require controls. Control of emissions from these hot metal processes to the proposed levels will require large increases in the volume of air being moved through the ventilation systems. Additional costs will be incurred to increase capacities of scrubbers and baghouses to remove the contaminants from the air.

The $71.0 million estimate includes engineering controls for processes where none are currently in use, as well as additional control of some already controlled processes. The controls for which costs have been estimated are sufficient for essentially all of the facilities in this sector. However, this estimate may somewhat underestimate the compliance costs because it does not take into account the additional costs at a small number of very large facilities where these engineering controls may not be sufficient. This situation arises in SIC 3312 at the blast furnaces and basic oxygen furnaces (BOFs) in the few (about 15) remaining integrated steel mills, which operate on a substantially larger scale than the other facilities in this sector. It is the scale of these approximately 15 operations (which account for about 80 percent of domestic steel production) which requires more extensive controls than other facilities in this sector.

Because engineering controls alone are likely to be insufficient to consistently control exposures to the proposed PELs around the blast furnace and BOFs at the integrated mills, OSHA anticipates that respirators will be needed in addition to the engineering controls. Engineering controls could possibly be installed to fully meet the proposed exposure levels, but the cost would likely be prohibitive, about $10 million per facility. The estimated cost of $71.0 million for this sector takes into account engineering controls such as improved air purification in control rooms and purified air showers at some work stations. These improvements would help to control exposures, but might not always be sufficient to meet the new standards. Thus, OSHA has also included an annual cost of $7.41 million for respirators at the integrated mills (included in the $71 million total estimate).

Fabricated Metal Products Manufacturing (SIC 34). Plating and coating establishments (SIC 347) and miscellaneous fabricated products (SIC 349) would account for a major portion of the $39.4 million annual costs in SIC 34. Worker exposures in this industry sector result from chemicals used in plating processes, solvents and coatings, metals and dusts. The survey indicated that ventilation systems are not now present at many of the processes with chemical exposure.

Machinery Except Electrical (SIC 35) and Electrical Machinery (SIC 36). The machinery manufacturing sectors together are estimated to incur total annual compliance costs of $65.9 million. Machinery except electrical accounts for $45.2 million of this total. The electrical machinery sector is estimated to require $20.7 million in annual compliance costs. Controls in these sectors would be required for exposures to metals, solvents and welding fumes.

Transportation Equipment Manufacturing (SIC 37). Annual costs of compliance for SIC 37 are estimated at $49.8 million. Costs in the truck and car body and motor vehicle parts sectors (SICs 3711, 3713, 3714) would account for a large percentage of the costs in SIC 37. Controls may be needed in order to control exposures to heavy metals, solvents, welding fumes and a large variety of other chemicals at large scale hot processes. Additionally, costs in small plants in this sector will include compliance activities to control exposures to styrene and other chemicals in small boat construction, as well as trailer and recreational vehicle insulation.

The Styrene Information Research Council (SIRC) presented the results of a study by Arthur D. Little, Inc. (ADL) of the costs of meeting a 50-ppm, 8-hour TWA for styrene in the boat-building industry [Exs. 3-742; 34A; Tr. 8/3/88, pp. 5-117 to 5-130]. The ADL study concluded that capital costs for boat builders would be $714.3 million and operating costs would be $132.1 million per year [Ex. 34A, Table 2]. If these costs are annualized using OSHA's interest and life-of-equipment assumptions, annualized costs for firms in this sector would be $249.2 million per year. OSHA's PRIA estimated a total annualized cost for all of SIC 37 (which includes many other segments in addition to boat-building) of $47 million per year (53 FR 21736). There is thus a substantial disagreement between ADL's chemical- and industry-specific estimate and OSHA's estimate for the entire sector.

OSHA believes that ADL's estimates grossly overestimate costs for controlling styrene in the boat-building sector. For example, ADL estimates that 50 percent of all workers in this sector are exposed to styrene levels of greater than 50 ppm as 8-hour TWA's. However, Daniel Boyd, testifying for SIRC, estimated that not more than 20 percent of employees engaged in boat building are directly exposed to styrene in the gel-coat and lamination processes; according to Dr. Boyd, the remainder of employees work in assembly and shipping and have little direct exposure to styrene [Tr. 8/3/88, p. 5-100]. OSHA's site visits to boat-building facilities in this sector [Exs. 136A, 136B] confirm that no more than 20 percent of employees in boat building facilities work in jobs having direct exposure to styrene.

ADL also chose to use 1980 exposure data to construct an exposure baseline for costs in this segment. OSHA finds the extensive exposure and control data collected in the Cal/OSHA study [Ex. 3-742, Attach. 2] superior to the 1980 ADL data because they are more recent, more extensive, specifically related to control measures (both engineering and work practice), and reflect good industrial hygiene practice (ADL, for example, calculates 8-hour TWAs on the basis of 1- or 2-hour samples, while Cal/OSHA uses appropriate sampling techniques). The Cal/OSHA study determined that the mean 8-hour TWA exposure for gel-coat and lamination workers (who are the most heavily styrene-exposed employees) were generally lower than reported by ADL.

In addition to overestimating both exposure levels and the number of workers overexposed, the ADL study [Ex. 33-742, Attach. 9, pp. 1-5] assumes that an extensive system of engineering controls and work practices would be required to achieve exposures of 50 ppm or less, i.e., ADL assumes a very low (or nonexistent) baseline level of control. However, both the Cal/OSHA study and OSHA's site visits [Exs. 136A, 136B] show that most gel-coat application is being done today in a spray booth [Ex. 3-742, Attach. 2, p. 20], and that many gel-coat operators have 8-hour TWA exposures of less than 50 ppm.

Further, based on the Agency's feasibility assessment for manual layup and sprayup operations within SIC 37, OSHA is permitting respirators to be used during these operations to achieve the revised limits for styrene. Thus, it is unlikely that employers will incur substantial costs to implement engineering controls for manual layup and sprayup operations.

Finally, ADL did not consider the impact of substitution of lower emitting styrene resins or of other, less hazardous substances in lieu of styrene on worker exposures. OSHA therefore concludes that the costs reflected in the PRIA for SIC 37, which are based on data from the survey and estimates developed by the cost algorithm, are an accurate representation of costs to firms in this sector.

Instruments Manufacturing (SIC 38). Annual control costs in SIC 38 are estimated to total $9.6 million. Exposures in this sector are to a large number of chemicals used within instruments and to various metals and solvents.

Miscellaneous Manufacturing (SIC 39). This industry accounts for a wide range of products, processes and chemical exposures. About half of the establishments that would incur the $15.8 million annual cost in the industry are believed to be included in SIC 3999, miscellaneous manufacturing not elsewhere classified.

The Casket Manufacturers Association of America (CMAA) commented that achievement of the proposed hardwood dust limit of 1 mg/m(3) would impose prohibitive costs on casket manufacturers [Ex. 8-78]. The CMAA presented estimates of the costs it anticipates as a result of the proposed limit; these costs were derived by estimating per-machine ventilation costs, multiplying this estimate by the number of machines per plant, and then multiplying by 18 plants [Ex. 8-78]. The CMAA estimated costs from a zero (no control) baseline and from an incremental baseline [Ex. 8-78]. Because the use of a zero-cost baseline is not appropriate when estimating potential compliance costs, OSHA has focused on the CMAA's incremental costs. According to the CMAA, total costs for 18 companies to achieve a 1 mg/m(3) limit for hardwood dust would be $1.32 million, or $73,000 per plant.

OSHA believes that the CMAA has seriously overestimated compliance costs. First, the final rule has adopted a 5 mg/m(3) limit for wood dust (the 2.5 mg/m(3) Western red cedar dust limit does not affect casket manufacturers because they do not use this wood). Second, the CMAA estimates assume that all machines in all facilities will need local exhaust ventilation, when in fact only a few machines would need to be engineered since only hand- and machine-finishing operations present an exposure problem, according to the CMAA [Ex. 8-87]. Finally, the recent exposure data collected and submitted by the CMAA show that, even under a worst-case scenario, seven of nine sample results were below the 5 mg/m(3) limit (see detailed discussion for SIC 39 in the Technological Feasibility section of the preamble). These exposure results demonstrate that most employees and operations are already below the final rule's 5 mg/m(3) limit and will therefore incur no costs.

Thus OSHA finds that the costs projected by the CMAA are unlikely to be incurred by hardwood casket manufacturers. The Agency's PRIA cost estimates for SIC 39 appear to be accurate and take into account costs of the magnitude likely to be encountered by these manufacturers.

One comment was received from a participant concerned about the costs of achieving the proposed limit for styrene in the manufacturing of diving boards, a business that is classified in SIC 3949, Sporting and Athletic Goods (nec). This commenter [Ex. 3-380] was of the opinion that the equipment changes and plant restructuring required to comply with the proposed limit would require a complete shut-down of affected facilities, and that this closure would result in such a substantial loss of revenue that economic feasibility would become an issue [Ex. 3-380].

In response, OSHA notes that a review of the record evidence has shown that the great majority of all exposure samples and reinforced plastics facilities potentially affected by the revised standard are already achieving compliance with this limit or are very close to doing so (see discussion of Technological Feasibility for SICs 30 and 37). In those few cases where compliance is not presently being achieved, OSHA has determined that improved work practices, such as having employees leave the booth when not engaged in manual layup operations and having them stand downwind, and making minor adjustments in ventilation will achieve the final rule's PEL. Thus OSHA finds that the cost impacts projected by this commenter [Ex. 3-380] are not likely to be incurred by diving board manufacturers.

Transportation and Utilities (SIC 40, 45, 47, and 49). The transportation and utilities sectors (SICs 40, 45, 47, and 49) include a large number of establishments. However, operations at Railroad (SIC 40), and Air Transport establishments (SIC 45) are subject to regulation by other Federal agencies in addition to OSHA. Consequently, the number of establishments which would incur costs to comply with the final standard are limited. For railroads, OSHA's standards normally apply to off-track operations. The estimated cost of compliance for SIC 40 is $1.1 million, while the cost for SIC 45 is $3.7 million.

Transportation Services Sector (SIC 47). The $3.8 million annual costs in SIC 47 will primarily be incurred in SIC 4789, transportation services not elsewhere classified. This sector includes establishments which provide incidental services such as cleaning railroad ballast and other rail car maintenance.

Electric, Gas and Sanitary Service Utilities (SIC 49). Annual costs in the utilities sectors are estimated to total $38.0 million. Costs would result from installation and improvement of controls necessary for activities such as boiler/furnace feed preparation in electric services, odorant addition by natural gas companies and water treatment and purification of water supplies.

Edison Electric Institute (EEI) estimates that in electric utility operations where exposures are intermittent in nature and limited in duration, engineering controls to reduce exposure would likely cost one to two million dollars per generating unit [Ex. 3-831]. However, intermittent activities such as boiler and precipitator cleaning would not require the installation of engineering controls, so these costs would not be incurred. During these intermittent activities, workers do have the option of wearing respirators. OSHA's cost estimate for SIC 49 does reflect costs to control exposures to coal dust generated in material handling operations.

The remaining cost estimate from EEI is $12-46 million per unit to modify the approximately 428 positive pressure boilers currently operating in the United States. EEI contends that if electric utilities lost their flexibility in using personal protective equipment to meet the proposed PELs, the boilers would have to be modified to reduce potential leaks of nitrogen dioxide and sulfur dioxide [Ex. 3-831]. OSHA believes that in most cases where overexposures might occur, they could be corrected by general ventilation or directed blowers and by correcting the most severe emission points. The prediction for such radical and costly modifications of power generating equipment does not appear to be well grounded.

The Interstate Natural Gas Association of America (INGAA) stated that costs to control "exposure to emissions from combustion sources that are ducted to ambient air" would be prohibitively costly [Ex. 3-739]. However, INGAA did not provide any specific explanations as to possible errors in OSHA's cost analysis. Thus OSHA did not have any additional evidence with which to compare its costs. After reviewing its methodology and survey data, OSHA concludes that the costs of compliance for the natural gas industry were adequately represented.

Wholesale Trade (SICs 50, 51). Costs in the wholesale trade sectors (SICs 50, 51), are estimated to total about $17.2 million annually. A large percentage of the total number of establishments which would incur costs to comply with the final rule are in SIC 5093, Scrap and Waste Materials, wholesale.

Several of the commenters who submitted data and information on the technological feasibility of achieving the Agency's proposed grain dust standard of 4 mg/m(3) in SIC 5153, Wholesale Trade and Grain and Field Beans, also expressed concern about the costs of compliance OSHA estimated for this sector in the PRIA. The PRIA estimated that approximately 10 percent of the grain elevators classified in SIC 5153 would incur costs to meet the proposed 4 mg/m(3) PEL and that the average per-elevator annualized costs would be $6,000 per year (Ex. 33). OSHA's estimates were based on data derived from the survey and calculated using the cost algorithm.

The National Grain and Feed Association (NGFA) presented a different estimate of the compliance costs that owners of grain elevators in this SIC category would incur (Ex. 3-752) to reach the proposed 4 mg/m(3) PEL. To derive its estimates, the NFGA used the following assumptions:

(1) All grain elevators processing wheat, oats, or barley will need pneumatic dust control systems and do not now have them;

(2) Eighty-seven percent of all grain elevators process wheat, oats, or barley;

(3) The costs of pneumatic dust control systems are those estimated by Booz Allen in a study done for OSHA in connection with the Agency's grain handling standard (inflated by 15 percent to convert them from 1984 to 1988 dollars.) Using these assumptions, the NGFA estimated total capital costs for all affected grain elevators at $1.9 billion. If these costs are annualized using OSHA's interest and life-of-capital-equipment assumptions and including an operating cost component calculated at 10 percent of capital costs, annualized costs would be $500 million per year, most of which reflect costs for country elevators. If the NFGA's estimated capital costs are used as a starting point, the average annualized per-elevator cost would be $41,125.

OSHA finds that the NGFA's estimates seriously overstate potential compliance costs for two principal reasons:

(1) OSHA has determined, as described for SIC 51 in the Technological Feasibility section of the preamble, that the PEL established in the final rule will be 10 mg/m(3), rather than the proposed PEL of 4 mg/m(3);

(2) The NGFA overestimates the number of grain elevators potentially affected by the new standard.

OSHA believes that no more than 10 percent, rather than the 87 percent projected by the NGFA, of all SIC 51 grain elevators will incur costs to achieve the 10 mg/m(3) PEL, because most elevators are already achieving this level. Data in the record show that:

(1) Only 5 percent of 109 8-hour TWA samples taken in grain elevators in one study were above 10 mg/m(3) [Rankin et al. 1986];

(2) Fewer than 5 percent of 203 8-hour TWA samples from grain handling facilities characterized as "small" were above 10 mg/m(3) [Ex. 3-751, Attach. 2 and Fig. 1, Docket H-0117];

(3) Only 12 percent of all total dust samples taken at 6 elevators in 3 states were above 10 mg/m(3) [Ex. 3-751, Attach., Docket H-0117]; and

(4) Only 6 percent of the employee full-shift exposures taken by NIOSH in a grain elevator were above 10 mg/m(3) [NIOSH HHE 76-13-316].

These data confirm that no more than 10 percent of all SIC 5153 elevators will be affected by the final standard. Further, these data make it clear that controls will be needed only in those instances and areas where the 10 mg/m(3) is not already being achieved, and that the complete, facility-wide installation of pneumatic control systems envisioned by the NGFA to meet a 4 mg/m(3) PEL will rarely, if ever, be required. OSHA has not reduced the compliance costs included in the PRIA for this sector despite the increase in the PEL from 4 to 10 mg/m(3); as such, costs are believed to be conservative and may overstate actual expenditures needed to comply with the new level.

Auto Dealers (SIC 55). The only retail trade sector expected to incur compliance costs, Auto Dealers (SIC 55) is estimated to incur $13.6 million annually. These costs result from the potentially large number of motor vehicle dealers (SIC 5511) which may incur compliance costs to control exposures to paints, coatings and solvents during vehicle spray and coating operations. The costs result from the installation of paint spray booths.

Service Sectors (SICs 72, 73, 75, 76, and 80). The service sectors, SICs 72, 73, 75, 76 and 80 are estimated to total about $26.7 million in annual compliance costs. The major costs in these sectors would result from potential compliance activities in SIC 721, laundry, cleaning and garment services. Establishments in SIC 721 would incur annual operating and annualized capital costs to control exposures for dry cleaning operations.

Because the limit for perchloroethylene was lowered from the proposed level of 50 ppm to 25 ppm, the engineering control designed for dry cleaning was reevaluated. OSHA reevaluated the control design used to project cost in the preliminary regulatory impact analysis. The air flow rate to control exposures at 25 ppm was increased, resulting in a unit cost increase of $910, making the revised unit cost $2,410. OSHA is aware of the improvements in dry cleaning equipment, particularly the increasing use of dry-to-dry machines. Based on information provided by the International Fabricare Institute and the Amalgamated Clothing and Textile Workers Union regarding replacement rates for drycleaning machines, OSHA believes that virtually all machines in use will be dry-to-dry by 1992 [Ex. 3-671]. The average perchloroethylene exposure associated with dry-to-dry machines is 23.9 ppm. Thus, it is anticipated that the PEL will be met largely by the normal rate of retirement of existing equipment.

Additional costs in the service sectors may result from control of solvent chemicals in SIC 734, Building Services, control of welding fumes at Welding Repair operations (SIC 7692), control of solvent and photographic chemicals in Mailing, Reproduction, Commercial Art, Photography and Stenographic Services (SIC 733), and local ventilation for exposure control in SIC 8071, Medical Laboratories.

Per Plant Average Costs

Table G-3 presents the estimated average per plant annual cost of compliance by industry sector. Costs shown in this Table are calculated only for those establishments in a sector which would incur costs. Average per plant annual operating and annualized capital costs for all affected establishments across industry sectors are estimated at $6,000. The per plant cost for large plants is $13,000 and for small plants with fewer than 20 employees, $3,100.


Table G-3:  AVERAGE PER PLANT ANNUAL COSTS AND NUMBERS OF AFFECTED
            PLANTS (a)

    NOTE:  Because of its width, Table G-3 has been divided; see
           continuation for additional columns.
_________________________________________________________________________
                                                                   AVERAGE
                                          TOTAL      # OF            COST/
                                           # OF  AFFECTED       %   AFFTED
SIC   SIC DESCRIPTION     ANNUAL COST    PLANTS    PLANTS  AFFTED    PLANT
(b)
__________________________________________________________________________

20   FOOD PROD. (c)       $33,493,100    29,000     4,932  16.98%   $6,800
21   TOBACCO (c)              $19,700       200         3   1.39%   $6,600
22   TEXT. MILL (c)       $29,748,400    11,000     2,765  25.08%  $10,700
23   APPAREL PROD. (c)    $31,744,200    30,000     6,179  20.57%   $5,100
24   LUMBER & WOOD        $56,720,800    27,100    18,427  68.00%   $3,100
25   FURNITURE            $21,075,800    12,700     5,062  40.00%   $4,200
26   PAPER PROD.          $30,998,700     7,000     3,518  50.00%   $8,800
27   PRINTING & PUB.      $33,754,500    60,300     3,597   6.88%   $9,400
28   CHEMICAL PROD.       $35,454,700    16,400     3,007  18.31%  $11,800
29   PETRO. REFINING      $23,686,000     2,300       306  13.25%  $77,400
30   RUBBER & PLASTICS   $111,093,400    15,100     3,562  26.22%  $31,200
31   LEATHER PROD.         $2,414,700     2,300       300  13.46%   $8,000
32   STONE & CLAY         $22,457,800    15,900     3,267  22.80%   $6,900
33   PRIM. METAL          $70,957,600     8,000     2,411  30.03%  $29,400
34   FAB. METALS          $39,419,700    37,300     4,597  14.50%   $8,600
35   MACHINERY            $45,206,600    64,400     6,801  10.56%   $7,800
36   ELEC. MACH.          $20,667,500    21,600     2,359  10.92%   $7,800
37   TRANS. EQUIP.        $49,792,400    13,600     4,979  36.56%  $10,000
38   INSTRUMENTS           $9,633,500    12,000     1,289  10.74%   $7,800
39   MISC. MANUF.         $15,842,600    25,300     2,649  10.47%   $7,800
40   R.R. TRANS.           $1,083,400       400        93  20.86%  $11,700
45   AIR TRANS.            $3,740,500     5,500       320   5.79%  $11,700
47   TRANS. SERV.          $3,789,400    26,200       324   1.24%  $11,700
49   ELEC. GAS. SAN.      $38,009,300    15,800     3,485  22.24%  $10,900
50   WHOLESALE TRADE       $2,995,300     5,800       801  13.78%   $3,400
51   WHOLESALE, NON-DUR   $14,215,800    33,600     4,436  13.22%   $3,400
55   AUTO DEALERS         $13,550,500   165,800    24,847  14.99%     $360
72   PERSONAL SRV.        $10,872,100    95,500     5,217   5.47%   $2,200
73   BUSINESS SRV.         $2,422,100    12,100       800   6.61%   $2,200
75   AUTO REPAIR           $6,143,500    91,500     8,351   9.13%     $600
76   MISC. REPAIR SRV.     $2,809,900    15,100     1,163  11.56%   $2,400
80   HEALTH SERV. (c)      $4,439,400   222,800     1,158   0.52%   $3,800
__________________________________________________________________________
TOTAL                    $787,982,900 1,101,600   131,005  11.89%   $6,000
__________________________________________________________________________

Table G-3:  AVERAGE PER PLANT ANNUAL COSTS AND NUMBERS
            OF AFFECTED PLANTS (a) - Continuation
______________________________________________________
                          AVERAGE COST    AVERAGE COST
                             PER LARGE       PER SMALL
SIC  SIC DESCRIPTION    AFFECTED PLANT  AFFECTED PLANT
------------------------------------------------------

20   FOOD PROD. (c)            $13,000          $3,600
21   TOBACCO (c)                $6,600              $0
22   TEXT. MILL (c)            $21,400          $3,700
23   APPAREL PROD. (c)         $11,500          $2,000
24   LUMBER & WOOD              $4,200          $2,700
25   FURNITURE                 $12,400          $1,800
26   PAPER PROD.               $15,200            $800
27   PRINTING & PUB.            $6,200         $10,600
28   CHEMICAL PROD.            $16,200          $5,400
29   PETRO. REFINING          $109,600            $700
30   RUBBER & PLASTICS         $27,000         $35,100
31   LEATHER PROD.             $10,400          $6,400
32   STONE & CLAY              $12,200          $3,400
33   PRIM. METAL               $41,900          $6,900
34   FAB. METALS               $15,800          $3,800
35   MACHINERY                 $14,600          $3,000
36   ELEC. MACH.               $14,500          $3,000
37   TRANS. EQUIP.             $11,800          $8,800
38   INSTRUMENTS               $14,500          $3,000
39   MISC. MANUF.              $14,600          $3,000
40   R.R. TRANS.               $11,700              $0
45   AIR TRANS.                $11,700              $0
47   TRANS. SERV.              $11,700              $0
49   ELEC. GAS. SAN.           $17,000          $3,600
50   WHOLESALE TRADE            $6,200          $2,900
51   WHOLESALE, NON-DUR         $6,200          $2,900
55   AUTO DEALERS               $2,000            $300
72   PERSONAL SRV.              $6,000          $1,000
73   BUSINESS SRV.              $8,300          $1,500
75   AUTO REPAIR                $3,500            $300
76   MISC. REPAIR SRV.         $12,400          $2,100
80   HEALTH SERV. (c)          $12,500          $2,100
______________________________________________________
TOTAL                         $ 13,000         $ 3,100

Source:  U.S. Department of Labor, Occupational Safety and Health
         Administration, Office of Regulatory Analysis

  Footnote(a):  Costs were calculated by annualizing the capital cost over
the projected life of the equipment (10 years) using a 10 percent cost of
capital and adding an annual operating and maintenance cost estimated at
10 percent of the capital cost.
  Footnote(b):  Industry sectors not identified in this table include
industries with no major cost impact expected, the construction industry,
which will be the subject of a separate regulatory analysis, and
industries such as mining, over which OSHA has no jurisdiction.
  Footnote(c):  Costs in these sectors were based on expert judgement and
secondary data collection.

The highest costs on an average per plant basis are expected to occur in SIC 29. Average per plant costs for large plants in SIC 29 may total $109,600 in annual operating and annualized capital costs. Per plant costs in SIC 29 are substantially higher than those in the next highest industry, SIC 30, Rubber and Plastics. The $31,200 per plant costs in this industry result from above average compliance costs estimated for exposure control in molding and vulcanizing in large plants and crushing and grinding operations in small plants.

Although small establishments account for about 73 percent of the 131,005 affected establishments, compliance costs for small establishments are expected to account for only 36 percent of total industry compliance costs.


Table G-4:  DESCRIPTIVE INFORMATION ON PROCESSES AND RELATED COSTS
__________________________________________________________________________

SIC 24 - LUMBER AND WOOD

                         FREQUENCY      TOTAL  COSTED           TOTAL COST
                                OF      WORK     WORK     UNIT         FOR
PROCESS                    PROCESS  STATIONS STATIONS     COST     PROCESS
__________________________________________________________________________

ABRASIVE SAND BLASTING         206         0        0   $7,200          $0
ADHESIVE GRINDING                2         4        4   $3,070     $12,115
ASSEMBLY                       237     3,567      172   $1,140    $196,170
BATCH PROCESS COKE
  PRODUCTION/REMOVAL            53         0        0       $0          $0
BLEACHING                       53       264      264   $2,900    $765,762
BOILERS                         57         6        0     $180          $0
CALENDARING/WINDING              2         4        0     $180          $0
CLEANING                       205         0        0     $710          $0
COATING/SPRAYING/FINISHING/
  LAYUP                      6,048    22,501    1,775   $3,070  $5,448,216
CRUSHING/GRINDING/CALCINING      2         0        0   $4,740          $0
CUTTING/SAWING/PLANING      21,525    96,614   15,458   $1,900 $29,370,656
DRYING/BAKING                1,085     3,834      181   $4,740    $858,878
GLUEING/HOT PRESSING         9,070    25,453    1,182   $3,070  $3,628,740
LOADING/OFFLOADING/
  RECEIVING/HANDLING             2        59        0   $1,120          $0
METAL WORKING (ROLLING,
  MILLING, SHAPING)              2       197        0   $1,140          $0
OTHER                          473    13,799        0   $1,140          $0
PLATE CLEANING                  42        77        0     $710          $0
POLISHING (SURFACE)/GRINDING     7        21        0   $1,140          $0
PULP SCREENING/WASHING           2        22       22   $2,900     $62,947
REBLENDING/REMIXING              2        22        0   $1,140          $0
RECOVERY/REPROCESSING/
  RECLAMATION                    2         4        0  $21,900          $0
SANDING/POLISHING/GRINDING   7,574    45,225    7,236   $2,200 $15,919,200
SEPARATION                       2         4        0   $1,120          $0
STAMPIMG/SHAPING/MOLDING/
  PRESSING                       9        53        0   $2,900          $0
WELDING/SOLDERING                2         4        0   $1,140          $0
ZSUBTOTAL                   46,664   211,734   26,294  $71,450 $56,262,684
ZZMAINTENANCE               18,116     2,743      148     $520    $457,965
ZZTOTAL                     64,780   214,477   26,442  $71,970 $56,720,649


SIC 25 - FURNITURE

                         FREQUENCY      TOTAL  COSTED           TOTAL COST
                                OF      WORK     WORK     UNIT         FOR
PROCESS                    PROCESS  STATIONS STATIONS     COST     PROCESS
__________________________________________________________________________

ADHESIVE BINDING                 5         5        0   $3,070          $0
ASSEMBLY                       216       489        0   $1,140          $0
COATING/SPRAYING/FINISHING/
  LAYUP                      6,117    13,182      685   $3,070  $2,102,014
CUTTING/SAWING/PLANING       6,821    28,050    4,460   $1,900  $8,474,469
DEBURRING                      229       457        0   $7,200          $0
DRILLING/BORING              1,040     1,197      192   $2,200    $421,344
DRYING/BAKING                  704     1,555        0   $4,740          $0
GLUEING/HOT PRESSING         5,432    10,829        0   $3,070          $0
LOADING/OFFLOADING
  RECEIVING/HANDLING            14       180       14   $1,120     $15,541
MACHINING/GRINDING/WELDING
  BRAZING                       51       771        0   $1,140          $0
OTHER                           56     1,169        0   $1,120          $0
PACKAGING/BAGGING               65        84        0   $1,120          $0
SANDING/POLISHING/GRINDING   7,539    28,052    4,488   $2,200  $9,874,304
STAMPING/SHAPING/MOLDING/
  PRESSING                      99       718        0   $2,900          $0
WELDING/SOLDERING               51       514        0   $1,140          $0
ZSUBTOTAL                   28,439    87,252    9,839          $20,887,672
ZZMAINTENANCE               12,804       945      361     $520    $187,885
ZZTOTAL                     41,243    88,197   10,200          $21,075,557

SIC 26 - PAPER PRODUCTS

                         FREQUENCY      TOTAL  COSTED           TOTAL COST
                                OF      WORK     WORK     UNIT         FOR
PROCESS                    PROCESS  STATIONS STATIONS     COST     PROCESS
__________________________________________________________________________

ADHESIVE BINDING                37         7        7   $3,070     $20,810
BLEACHING                       38       150       73   $2,900    $211,326
BLENDING/MIXING/FORMULATING    135       363        0   $3,070          $0
BOILERS                         38       119       11     $180      $2,033
CALENDARING/WINDING            213       739      136     $180     $24,549
COATING/SPRAYING/FINISHING/
  LAYUP                        555     2,883      141   $3,070    $433,631
CRUSHING/GRINDING/CALCINING      2         5        0   $4,740          $0
CUTTING/SAWING/PLANING       1,899     9,504    1,521   $1,900  $2,889,216
DIGESTER                        29       138        9  $14,000    $126,535
DON'T KNOW                     756         0        0                   $0
DRYING/BAKING                1,328     9,052      305   $4,740  $1,445,978
EXTRUSION                       71       140        2   $1,140      $2,575
FLEXOGRAPHIC PRINTING            2        43        0   $1,380          $0
FOAM PROCESSING                 69     1,718        0   $1,140          $0
GLUEING/HOT PRESSING         1,663     8,384      873   $3,070  $2,679,276
LINOTYPE SETTING                 2         5        0   $1,380          $0
LITHOGRAPHIC PRINTING            7        23        0   $1,380          $0
LOADING/OFFLOADING
  RECEIVING/HANDLING             5         9        0   $1,120          $0
OTHER                          254     1,248        0      $90          $0
PACKAGING/BAGGING            1,669    12,651    1,452   $1,120  $1,625,905
PRESS SECTION                  246       962      136     $180     $24,492
PRINTING                       716     3,666      946   $1,380  $1,305,073
PULP SCREENING/WASHING          38       237      133   $2,900    $386,612
RECOVERY/REPROCESSING/
  RECLAMATION                   38        91       25  $14,000    $352,492
SHEET PROCESS                    2         7        0      $90          $0
SHREDDING/WASTE PROCESSING   1,347     4,309    1,198  $14,000 $16,774,349
SIZE PRESS/COATERS             366       990      611     $180    $109,969
STAMPING/SHAPING/MOLDING/
  PRESSING                      46       132       11   $2,900     $32,763
WATER TREATMENT                 67       122       74  $14,000  $1,031,658
WET END                        233       753      182     $180     $32,802
ZSUBTOTAL                   11,871    58,370    7,847          $29,512,044
ZZMAINTENANCE                7,022     3,357    2,858     $520  $1,486,291
ZZTOTAL                     18,893    61,727   10,705          $30,998,335

SIC 27 - PRINTING AND ALLIED INDUSTRIES

                         FREQUENCY      TOTAL  COSTED           TOTAL COST
                                OF      WORK     WORK     UNIT         FOR
PROCESS                    PROCESS  STATIONS STATIONS     COST     PROCESS
__________________________________________________________________________


ADHESIVE BINDING             5,567     8,109      197   $3,070    $604,003
BLENDING/MIXING/FORMULATING     39       197        0   $4,740          $0
BOILERS                         20         0        0     $180          $0
CALENDARING/WINDING            113       678        0     $180          $0
CLEANING                     1,602     9,629    9,629     $710  $6,836,424
CUTTING/SAWING/PLANING          38       283        0     $180          $0
DEGREASING                      20        20        0     $710          $0
DRILLING/CUTTING/FLAME-JET/
  LANCING                       18       111      111   $1,140    $126,499
DRYING/BAKING                   39        39        0   $4,740          $0
FILM PROCESSING              2,115     4,194        0   $1,240          $0
GRAVURE PLATEMAKING            361     1,105        0   $1,380          $0
GRAVURE PRINTING                22       328        0   $1,380          $0
INJECTION MOLDING               20       315        0   $1,140          $0
LETTERPRESS PRINTING         7,526    18,729        0   $1,380          $0
LINOTYPE SETTING             8,372    12,642       20   $1,380     $27,150
LITHOGRAPHIC PLATEMAKING    17,757    22,169       99   $1,380    $136,716
LITHOGRAPHIC PRINTING       27,040    87,625   17,540   $1,380 $24,204,521
MATERIALS MANUFACTURE/
  FABRICATION                   20       236        0   $1,140          $0
METAL PLATING                   20       138        0     $710          $0
MONO OR LINOTYPE SETTING     6,367     9,629        0   $1,380          $0
OTHER                        3,884     4,900        0   $1,140          $0
PHOTOENGRAVING                 418     1,132        0   $1,380          $0
PHOTOENGRAVING PLATEMAKING     113         0        0   $1,380          $0
PHOTOGRAVURE                 1,850     5,415      678   $1,380    $936,090
PLATE CLEANING                 113         0        0     $710          $0
PLATE MAKING                   268       764        0   $1,380          $0
PRINTING                     3,687     6,981        0   $1,380          $0
SANDING/POLISHING/GRINDING      20        20        0   $2,200          $0
SCREEN PRINTING                304     1,306        0   $1,380          $0
SCREEN STENCIL PLATEMAKING     133       585        0   $1,380          $0
SEMICONDUCTOR PHOTORESIST       20       413        0     $710          $0
STAMPING/SHAPING/MOLDING/
  PRESSING                      18       148      148   $2,900    $429,063
STRIPPING/PAIN REMOVING        113       452        0   $1,140          $0
ZSUBTOTAL                   88,114   198,311                   $33,300,466
ZZMAINTENANCE               60,282    19,919      678     $520    $454,041
ZZTOTAL                    148,396   218,230   29,099          $33,754,507

SIC 28 - CHEMICALS AND ALLIED PRODUCTS

                         FREQUENCY      TOTAL  COSTED           TOTAL COST
                                OF      WORK     WORK     UNIT         FOR
PROCESS                    PROCESS  STATIONS STATIONS     COST     PROCESS
__________________________________________________________________________

ADHESIVE BINDING               104       219        0   $3,070          $0
ASSEMBLY                         6       714        0   $1,140          $0
BLENDING/MIXING/FORMULATING  7,640    35,776    1,988   $3,070  $6,104,570
BLOWING/MOLDING                  2         6        0      $90          $0
BOILERS                          7        89        0     $180          $0
CALCINING KILN                   3         5        0   $4,740          $0
CALENDARING/WINDING              8       604      604     $180    $108,707
CLEANING                        71       471       26     $710     $18,119
COATING/SPRAYING/FINISHING/
  LAYUP                        137       570       16   $3,070     $47,818
CRUSHING/GRINDING/CALCINING  1,196     6,740      542   $4,740  $2,570,110
CUTTING/SAWING/PLANING          80       149        0     $180          $0
DIGESTER                        45       360        0  $14,000          $0
DRYING/BAKING                1,204     3,455      230   $4,740  $1,088,607
EXTRUSION                      222     1,460      123   $1,140    $140,758
GLUEING/HOT PRESSING             2         9        0   $3,070          $0
IMPREGNATION                    53       153        0   $1,140          $0
INJECTION MOLDING                2        28        0   $1,140          $0
LITHOGRAPHIC PRINTING            6         6        0   $1,380          $0
LOADING/OFFLOADING/RECEIVING/
  HANDLING                   6,100    14,631    2,065   $1,120  $2,312,937
MATERIALS MANUFACTURE/
  FABRICATION                    7        35        0   $1,140          $0
MEASUREMENT                     11        33        0   $1,120          $0
MONOTYPESETTING                 25        25        0   $1,380          $0
OTHER                          162     6,899        0   $1,120          $0
OTHER                            3        15        0      $90          $0
OTHER                          261       392        0      $90          $0
OTHER                            3         0        0     $710          $0
PACKAGING/BAGGING            5,137    23,614    4,538   $1,120  $5,082,081
PACKAGING/REPACKAGING           29        29        0     $710          $0
PRINTING                         2        14        0   $1,380          $0
PROCESS INSP/SUPERVISION/
  QUAL CONTROL                  39        80        0   $1,120          $0
PULP SCREENING/WASHING           5       638        0   $2,900          $0
REACTION/FERMENTATION        1,680     7,889    3,855   $1,120  $4,317,769
RECOVERY/REPROCESSING/
  RECLAMATION                  628     1,113      429  $21,900  $9,402,121
SAMPLING                         2         0        0   $1,120          $0
SANDING/POLISHING/GRINDING      36        69        0   $2,200          $0
SEPARATION                   1,521     6,026    2,269   $1,120  $2,541,828
STAMPING/SHAPING/MOLDING/
  PRESSING                      45       169       20   $2,900     $58,846
STERILIZATION                    8         8        6     $710      $4,225
VULCANIZATION/CURING            36       380        0      $90          $0
WATER TREATMENT                 13        38       37  $14,000    $511,930
ZSUBTOTAL                   26,539   111,911   16,748          $34,310,426
ZZMAINTENANCE               16,426     3,952      497     $520  $1,144,315
ZZTOTAL                     42,965   115,863   17,245          $36,454,741

SIC 29 - PETROLEUM REFINING AND RELATED PRODUCTS

                         FREQUENCY      TOTAL  COSTED           TOTAL COST
                                OF      WORK     WORK     UNIT         FOR
PROCESS                    PROCESS  STATIONS STATIONS     COST     PROCESS
__________________________________________________________________________

ADHESIVE BINDING                 8        15        0   $3,070          $0
BATCH PROCESS COKE
  PRODUCTION/REMOVAL            47       208       15 $150,000  $2,281,338
BLENDING/MIXING/FORMULATING    649     2,004       23   $4,740    $107,388
CALCINING KILN                   2        11        0   $4,740          $0
CALENDARING/WINDING             11        13        0     $180          $0
COATING/SPRAYING/FINISHING/
  LAYUP                         24        38        0   $3,070          $0
CUTTING/SAWING/PLANING          12        14        2     $180        $392
DRYING/BAKING                  108       178        0   $4,740          $0
FELTING                          2         2        0   $3,070          $0
IMPREGNATION                     6         6        0   $3,070          $0
LOADING/OFFLOADING/RECEIVING/
  HANDLING                     745     4,065      278   $1,120    $311,593
MEASUREMENT                    112       787        0   $1,120          $0
OTHER                            2         4        0                   $0
OTHER                           11        45        0   $1,120          $0
OTHER                            2         9        0   $4,740          $0
OTHER                            6         0        0   $4,740          $0
OTHER                           45       116        0   $4,740          $0
PACKAGING/BAGGING              342       931       83   $1,120     $92,457
PRESS SECTION                    2         2        0     $180          $0
PROCES INSP/SUPERVISION/
  QUAL CONTROL                 214     2,299    1,441   $1,120  $1,614,469
REACTION/FERMENTATION           37       205       72   $1,120     $80,102
RECOVERY/REPROCESSING/
  RECLAMATION                   25        54        0  $21,900          $0
SAMPLING                       267     5,790    3,065   $1,120  $3,433,000
SAMPLING OF PIPE LINES          16        16        0   $1,120          $0
SANDING/POLISHING/GRINDING       2         7        0   $2,200          $0
SEPARATION                      41        73       28   $1,120     $30,912
WATER TREATMENT                215     1,646    1,103  $14,000 $15,440,626
WELDING/SOLDERING                2         0        0   $1,140          $0
ZSUBTOTAL                    2,953    18,538    6,109          $23,392,277
ZZMAINTENANCE                2,315       875      318     $520    $293,765
ZZTOTAL                      5,268    19,412    6,427          $23,686,042

SIC 30 - RUBBER AND MISCELLANEOUS PLASTIC PRODUCTS

                         FREQUENCY      TOTAL  COSTED           TOTAL COST
                                OF      WORK     WORK     UNIT         FOR
PROCESS                    PROCESS  STATIONS STATIONS     COST     PROCESS
__________________________________________________________________________


ABRASIVE BLASTING               17        17        0   $7,200          $0
ASSEMBLY                        35       653       32   $1,140     $36,352
BLENDING/MIXING/FORMULATING  4,537    10,538      188   $3,070    $577,615
BONDING                          6         6        0   $3,070          $0
CALENDAR/WINDING             1,375     4,398      144     $180     $25,905
CASTING                         21       469        0   $3,890          $0
CLEANING                        19       243        0     $710          $0
COATING/SPRAYING/FINISHING/
  LAYUP                      1,730    15,531   11,224   $3,070 $34,457,310
CONFINED SPACE EXPOSURE TO
  EXHAUST FUME                 107       269        0      $80          $0
CRUSHING/GRINDING/CALCINING  1,360     2,005       54   $4,740    $254,597
CUTTING/SAWING/PLANING          52       382        0     $180          $0
DEGREASING                   2,331     2,334    2,220     $710  $1,575,895
DIGESTER                        48        48        0  $14,000          $0
ELECTROPLATE/ELECTRICAL
  DISCHARGE MACHINE             16        16        0     $710          $0
EXTRUSION                      794     4,322      176   $1,140    $200,590
FOAM PROCESSING                746    14,587       10   $1,140     $10,970
GLUEING/HOT PRESSING         3,394    21,705    2,331   $3,070  $7,155,104
INJECTION MOLDING              245     1,204      973   $1,140  $1,109,058
LOADING/OFFLOADING/
  RECEIVING/HANDLING         1,137     2,689      193   $1,120    $215,850
MACHINING/GRINDING/WELDING/
  BRAZING                       33       387      112   $1,140    $127,887
METAL PLATING                  105       105       16     $710     $11,185
METAL WORKING (ROLLING,
  MILLING, SHAPING)             17        17        0   $1,140          $0
OTHER                           57     1,016        0   $3,070          $0
PACKAGING/BAGGING              249       719       13   $1,120     $14,286
PRINTING                         3         6        0   $1,380          $0
REACTION/FERMENTATION            5         5        0   $1,120          $0
REBLENDING/REMIXING             38       304       61   $3,070    $185,786
SCREEN PRINTING                 16       158        0   $1,380          $0
SEPARATION                       6        26        0   $1,120          $0
STAMPING/SHAPING/MOLDING/
  PRESSING                   4,064    51,203   22,195   $2,900 $64,366,057
VULCANIZING/CURING             461     9,388       92   $3,070    $281,245
WATER TREATMENT                178       188        0  $14,000          $0
ZSUBTOTAL                   23,205    14,957   40,031         $110,605,792
ZZMAINTENANCE               15,094       632      107     $520    $486,138
ZZTOTAL                     38,299   145,589   40,138         $111,091,930

SIC 31 - LEATHER AND LEATHER PRODUCTS

                         FREQUENCY      TOTAL  COSTED           TOTAL COST
                                OF      WORK     WORK     UNIT         FOR
PROCESS                    PROCESS  STATIONS STATIONS     COST     PROCESS
__________________________________________________________________________


ASSEMBLY                       195       770        0   $1,140          $0
BEAMHOUSE                       29       171        0   $2,510          $0
BLENDING/MIXING/FORMULATING     12        35       35   $3,070     $61,419
CALENDARING/WINDING              12        12       47     $180
$8,375
CLEANING                        22       111       30     $710     $21,070
COATING/SPRAYING/FINISHING/
  LAYUP                        204       547        0   $3,070          $0
COLORING/DYEING                 44       420        0       $0          $0
CUTTING/SAWING/PLANING          59       232        0     $180          $0
DEGREASING                      12        91        0     $710          $0
EXTRUSION                       12       105        0   $1,140          $0
FAT LIQUORING                   27       281        0     $720          $0
FINISHING                       66       148        0   $1,820          $0
GLUEING/HOT PRESSING         1,509     5,540      736   $3,070  $2,258,853
INJECTION MOLDING               10        69        0   $1,140          $0
LOADING/OFFLOADING/
  RECEIVING/HANDLING            35        54        0   $1,120          $0
OTHER                           12       931        0   $1,120          $0
PACKAGING/BAGGING               35        35        0   $1,120          $0
PRESERVATION/DEFESTATION/
  DISINFECTION                  34        58       41     $720     $29,466
SPLITTING/SHAVING               22        77        0     $720          $0
STAMPING/SHAPING/MOLDING/
  PRESSING                      22       241        0   $2,900          $0
TANNING/RETANNING               51       158       10   $2,510     $25,887
ZSUBTOTAL                    1,971    10,120      999           $2,652,183
ZZMAINTENANCE                2,326         0        0     $520      $9,646
ZZTOTAL                      4,297    10,120      999           $2,414,719

SIC 32 - STONE, CLAY, GLASS AND CONCRETE PRODUCT MANUFACTURING

                         FREQUENCY      TOTAL  COSTED           TOTAL COST
                                OF      WORK     WORK     UNIT         FOR
PROCESS                    PROCESS  STATIONS STATIONS     COST     PROCESS
__________________________________________________________________________

ANNEALING/QUENCH/TEMPER         92       641        0     $710          $0
ASSEMBLY                        30       281        0   $1,140          $0
BATCH MIXING                 1,205     1,262        0   $4,740          $0
BISCUIT FIRING                 110       110        0   $4,740          $0
BLENDING/MIXING/FORMULATING  4,005    22,404      284   $4,740  $1,345,994
BLOWING/MOLDING                112     1,572      544      $90     $48,937
BONDING                      1,157     2,463    2,200   $3,070  $6,753,534
CALCINING KILN                 139       241       11   $4,740     $54,320
CALENDARING/WINDING             11        92        0     $180          $0
CASTING                      1,319     1,756      382   $3,890  $1,484,491
CHIPPING GRINDING            1,466     2,108      275   $1,140    $313,322
COATING/ETCHING                226       645       11   $3,070     $35,182
COATING/SPRAYING/FINISHING/
  LAYUP                        298     1,512       92   $3,070    $281,257
COLD ROLLING MILL               92       183        0   $1,820          $0
CRUSHING/GRINDING/CALCINING    900     2,033    1,168   $4,740  $5,538,889
CUTTING/SAWING/PLANING       1,349     3,618    1,269     $180    $228,376
DECORATION                     133       810       11     $710      $8,136
DRILLING/BORING                 18        73       73   $2,200    $159,501
DRILLING/CUTTING/FLAME-JET
  LANCING                    1,466     2,474        0   $1,140          $0
DRYING/BAKING                  341       636       54   $4,740    $257,739
EXTRUSION                      183       550        0   $1,140          $0
FIBER FORMING                   30       138        0      $90          $0
FINISHING                      110     1,924    1,924   $1,820  $3,501,522
GLAZE APPLICATION              187       276       91   $3,070    $278,220
GLOSS FIRING                   164       200       36   $4,740    $171,826
IMPREGNATION                    11       115        0   $1,140          $0
INJECTION MOLDING               11       138        0   $1,140          $0
LOADING/OFFLOADING/RECEIVING/
  HANDLING                     241       558       23   $1,120     $25,670
MACHINING/GRINDING/WELDING/
  BRAZING                      110     1,134        0   $1,140          $0
MATERIALS MANUFACTURE/
  FABRICATION                   18         0        0   $1,140          $0
MELTING                        249     1,256       36   $3,890    $141,013
METAL MELTING/POURING/CASTING   11        92        0   $1,820          $0
METAL PLATING                   30       229        0     $710          $0
MOLDMAKING                      11        57        0   $2,520          $0
OTHER                          110       182        0      $90          $0
PACKAGING/BAGGING              352       958       23   $1,120     $25,670
PACKAGING/REPACKAGING           11        23        0   $1,820          $0
POLISHING (SURFACE)/GRINDING 1,466     1,833        0   $1,140          $0
SANDING/POLISHING/GRINDING      18         0        0   $2,200          $0
SCREEN PRINTING                238       695        0   $1,380          $0
SCREEN STENCIL PLATEMAKING      18        36        0   $1,380          $0
SIZING                         157       673      127   $4,740    $601,392
SLIP HOUSE (BLENDING)          341       341        0   $1,140          $0
STAMPING/SHAPING/MOLDING/
  PRESSING                      82       749      138   $2,900    $398,807
VULCANIZATION/CURING            11        57        0      $90          $0
WELDING/SOLDERING              201       769      403   $1,140    $459,089
ZSUBTOTAL                   18,843    57,897   10,863          $22,112,897
ZZMAINTENANCE               15,920     8,620      344     $520    $344,949
ZZTOTAL                     34,763    66,516   11,207          $22,457,846

SIC 33 - PRIMARY METAL MANUFACTURING

                         FREQUENCY      TOTAL  COSTED           TOTAL COST
                                OF      WORK     WORK     UNIT         FOR
PROCESS                    PROCESS  STATIONS STATIONS     COST     PROCESS
__________________________________________________________________________

ABRASIVE BLASTING            1,504     4,674      560   $7,200  $4,035,395
ACID WASHING                    78       103        0     $710          $0
ANNEALING/QUENCH/TEMPER      1,381     7,106    1,315     $710    $933,430
BLENDING/MIXING/FORMULATING     12        12        0   $3,070          $0
CALCINING KILN                  10        10       10   $4,740     $46,198
CLEANING                        18        18        0     $710          $0
COATING & DRAWING              287       389        0   $7,200          $0
COATING/SPRAYING/FINISHING/
  LAYUP                         31       283        0   $3,070          $0
COKE MANUFACTURE                22        35        0       $0          $0
COLD ROLLING MILL              403     2,620      391   $1,820    $712,061
COLORING/DYEING                 18        37        0       $0          $0
CORE MAKING                    945     3,466      945   $2,620  $2,380,446
COATING/PAINTING               192       628        0   $3,070          $0
DEGREASING                     320       453       41     $710     $28,828
DEMAGGING                       28        48       24   $1,140     $27,164
ELECTROD PRODUCTION             37       100       50   $1,820     $90,555
ELECTROPLATE/ELECTRICAL
  DISCHARGE MACHING            109     6,737        0     $710          $0
EXTRUSION                       10        39        0   $1,140          $0
EXTRUSION COATING              236     1,325      297   $7,200  $2,139,242
FINISHING                      748     5,746      656   $1,820  $1,193,680
FORGING PRESS                   10         0        0   $1,820          $0
GLUEING/HOT PRESSING            94       188        0   $3,070          $0
HOT DIP GALVANIZING             12        12        0     $710          $0
HOT METAL WORKING               28       112       19   $7,200    $140,348
HOT SHAPING                    342     1,003      109   $2,520    $274,999
IMPREGNATION                   180       728      150   $1,140    $171,537
MACHINING/GRINDING/WELDING/
  BRAZING                      508     1,479      305   $1,140    $347,508
MAINTENANCE ACTIVITIES          12        25        0   $7,200          $0
MELTING                         77       192        0   $3,890          $0
METAL CASTING                1,299     5,649      686   $2,520  $1,727,898
METAL MELTING/POURING/
  CASTING                    3,522    14,024    2,767   $1,820  $5,036,591
METAL WORKING (ROLLING,
  MILLING, SHAPING)             74       145        0   $1,140          $0
MOLDMAKING                     556     2,170      145   $2,520    $365,710
ORE HANDLING                   110    19,631   19,589   $1,820 $35,651,199
OTHER                           89       387        0   $4,740          $0
PC BOARDS  SOLDERING            25        62       50   $1,140     $56,721
PICKELING                      357       749        5     $710      $3,743
POLISHING (SURFACE)/GRINDING 1,532     4,534    1,668   $1,140  $1,901,789
POURING                        608     3,298      131   $1,820    $239,227
PRESSING                        38     1,382        0   $3,070          $0
RAW MATERIALS PREPARATION      115       230        0   $2,520          $0
SAMPLING                        12       149        0   $1,120          $0
SAND RECLAMATION               115       152        0   $2,520          $0
SANDING                         47        47        0   $1,140          $0
SEPARATION                      28       139      203   $1,120    $226,914
SHAKEOUT                       832     3,337      468   $2,520  $1,179,773
SINTERING                      166       412       85  $21,900  $1,865,656
SOLDERING                       47        94        0   $1,140          $0
STAMPING/SHAPING/MOLDING/
  PRESSING                     102       626        0   $2,900          $0
STRIP ANNEALING                 94       939      469   $1,820    $854,157
TORCH CUTTING                  131       734       56   $2,520    $140,453
WELDING/BRAZING                 65       214      141   $1,140    $160,506
WELDING/SOLDERING              285       888      237   $1,140    $270,505
WIRE DRAWING                   177     1,917      289   $1,140    $329,715
ZSUBTOTAL                   18,083    99,477   31,862          $62,531,948
ZZMAINTENANCE               64,224    54,448    1,784     $520  $8,425,657
ZZTOTAL                     26,111   106,283   33,646          $70,957,605

SIC 34 - FABRICATED METAL PRODUCTS MANUFACTURING

                         FREQUENCY      TOTAL  COSTED           TOTAL COST
                                OF      WORK     WORK     UNIT         FOR
PROCESS                    PROCESS  STATIONS STATIONS     COST     PROCESS
__________________________________________________________________________

ABRASIVE BLASTING            3,523     3,957      138   $7,200    $993,370
ACID WASHING                 1,175     2,679        0     $710          $0
ANNEALING/QUENCH/TEMPER        609       559        0     $710          $0
ASSEMBLY                     1,189     7,018      618   $1,140    $704,382
BISCUIT FIRING                  24         0        0   $4,740          $0
BLENDING/MIXING/FORMULATING    178     1,069        0   $2,200          $0
COATING & DRAWING              177       148        0   $7,200          $0
COATING/SPRAYING/FINISHING/
  LAYUP                      1,119       940        0   $3,070          $0
COLD ROLLING MILL              144     2,290        0   $1,820          $0
COATING/PAINTING            12,164    22,825    1,411   $3,070  $4,332,852
CUTTING/SAWING/PLANING         422     2,139        0     $180          $0
DEBURRING                      129       271      247   $7,200  $1,779,493
DEGREASING                   6,315    15,464       55     $710     $39,384
DRILLING/BORING                406       406        0   $2,200          $0
DRILLING/CUTTING/FLAME-JET
  LANCING                       24        97        0   $1,140          $0
ELECTROD PRODUCTION             89        89        0   $1,820          $0
ELECTROPLATE/ELECTRICAL
  DISCHARGE MACHING          3,103    18,617    1,505     $710  $1,068,543
ENGRAVING/ETCHING              406       812        0     $710          $0
EXTRUSION COATING               62       510        0   $7,200          $0
FINISHING                      501       958        0   $1,820          $0
FORGING PRESS                   15       463      463   $1,820    $843,405
HOT DIP GALVANIZING            466     5,871       93     $710     $65,804
HOT METAL WORKING               89       891        0   $7,200          $0
HOT SHAPING                    240       926        0   $2,520          $0
LITHOGRAPHIC PRINTING           15       124      124   $1,380    $170,534
LOADING/OFFLOADING/RECEIVING/
  HANDLING                      89       356        0   $1,120          $0
MACHINING/GRINDING/WELDING/
  BRAZING                    5,051    29,045    1,219   $1,140  $1,389,146
MAKING OF DENTAL APPLIANCES     15       463        0   $1,140          $0
METAL MELTING/POURING/CASTING   15        46        0   $1,820          $0
METAL WORKING (ROLLING,
  MILLING, SHAPING)          1,363     6,551    3,656   $1,140  $4,167,440
MOLDMAKING                      89       267        0   $2,520          $0
OTHER                          502     1,691        0   $4,740          $0
PACKAGING/BAGGING              178       624        0   $1,120          $0
PICKELING                       64       112        0     $710          $0
POLISHING (SURFACE)/GRINDING 8,500    23,188    2,192   $1,140  $2,498,617
PRESSING                     2,294    15,729        0   $3,070          $0
PROCESS INSP/SUPERVISION/
  QUAL CONTROL                 446     1,280      121   $1,120    $135,841
REACTION/FERMENTATION           15        46        0   $1,120          $0
RECEIVING/SHIPPING              89       356        0   $1,120          $0
SANDING/POLISHING/GRINDING      89        89        0   $2,200          $0
SOLDERING                       24        97        0   $1,140          $0
STAMPING/SHAPING/MOLDING/
  PRESSING                   1,343    15,761    1,236   $2,900  $3,583,702
TORCH CUTTING                   89       891        0   $2,520          $0
WATER TREATMENT                 24        24       24  $14,000    $339,603
WELDING/SOLDERING           11,788    46,728   13,256   $1,140 $15,111,924
WIRE DRAWING                   247     2,605        0   $1,140          $0
ZSUBTOTAL                   64,903   235,074   26,358          $37,224,040
ZZMAINTENANCE               37,315    13,302    1,698     $520   2,195,645
ZZTOTAL                    102,218   248,377   28,056          $39,419,685

SIC 35, 36, 38, 39 - MACHINERY, INSTRUMENTS, AND MISC. MANUFACTURING

                         FREQUENCY      TOTAL  COSTED           TOTAL COST
                                OF      WORK     WORK     UNIT         FOR
PROCESS                    PROCESS  STATIONS STATIONS     COST     PROCESS
__________________________________________________________________________

ABRASIVE BLASTING           12,863    13,588      407   $7,200  $2,928,488
ACID WASHING                 2,809     2,809        0     $710          $0
ASSEMBLY                     5,841    64,324    5,054   $1,140  $5,762,058
BLENDING/MIXING/FORMULATING  1,573    11,016      731   $3,070  $2,243,008
BRISTLE/FIBER CLEANING/
  RECEIVING                     49        49        0   $1,140          $0
CLEANING                     2,015     3,071        0     $710          $0
COATING/SPRAYING/FINISHING/
  LAYUP                      7,478    19,943    1,162   $3,070  $3,567,133
COIL PRODUCTION                 49        49        0   $2,520          $0
CORE MAKING                    203     1,627        0   $2,520          $0
COATING/PAINTING            20,560    45,995    2,052   $3,070  $6,300,780
CRUSHING/GRINDING/CALCINING    314     1,905        0   $4,740          $0
CUTTING/SAWING/PLANING       6,902    25,579    1,017     $180    $183,030
DEGREASING                  17,198    28,737    1,927     $710  $1,367,926
DRILLING/BORING              2,886    26,868    2,034   $2,200  $4,474,080
DRYING/BAKING                   49        98        0   $4,740          $0
ELECTROPLATE/ELECTRICAL
  DISCHARGE MACHINING          877       877        0     $710          $0
EPOXY COATING                  456       407        0   $3,070          $0
EXTRUSION COATING              203       813        0   $7,200          $0
FILM & PRINT PAPER MAKING       49       196        0   $1,240          $0
FINISHING                       81       101        0   $1,820          $0
FOAM PROCESSING                 61        61       61   $1,140     $69,828
GLASSMAKING                    203       203        0      $90          $0
GLUEING/HOT PRESSING         6,672    12,260    4,871   $3,070 $14,953,587
HOT SHAPING                     27        27        0   $2,520          $0
INJECTION MOLDING            1,763     9,317        0   $1,140          $0
LOADING/OFFLOADING/
  RECEIVING/HANDLING         1,272     2,014        0   $1,120          $0
MACHINING/GRINDING/WELDING/
  BRAZING                    6,176    65,357      959   $1,140  $1,092,764
MATERIALS MANUFACTURE/
  FABRICATION                3,090    29,519    6,799   $1,140  $7,750,921
METAL MELTING/POURING/
  CASTING                    1,321     3,862    1,419   $1,820  $2,582,353
METAL WORKING (ROLLING,
  MILLING, SHAPING)          2,876     6,710        0   $1,140          $0
MOLDMAKING                     375     3,708    2,361   $2,520  $5,949,315
MONO OR LINOTYPE SETTING       110       110        0   $1,380          $0
ORE HANDLING                   203       203        0   $1,820          $0
OTHER                        1,301    15,739        0      $90          $0
PACKAGING/BAGGING            1,740     5,971      203   $1,120    $227,771
PAINTING/COATING                49         0        0   $3,070          $0
PC BOARDS - ETCHING            486       645      110   $2,520    $277,652
PC BOARDS - SOLDERING        1,370     1,494        0   $1,140          $0
PHOTOFINISHING                 252       960        0   $1,240          $0
PICKELING                      301       554        0     $710          $0
PLATE PROCESS                  265     1,446        0      $90          $0
POLISHING (SURFACE)/
  GRINDING                  31,073   108,851    1,728   $1,140  $1,969,642
POTTING                      1,136     2,880        0   $2,520          $0
PRESSING                     8,518    36,907    3,182   $3,070  $9,770,028
PRINTING                       380     1,268        0   $1,380          $0
PROCESS INSP/SUPERVISION/
  QUAL CONTROL               1,463     6,382        0   $1,120          $0
REACTION/FERMENTATION          456     2,132        0   $1,120          $0
RECEIVING/SHIPPING              61        61        0   $1,120          $0
REFRIGERANT CHARGING           203         0        0   $2,520          $0
SAMPLING                        49       489        0   $1,120          $0
SANDING                      3,829     4,820        0   $1,140          $0
SANDING/POLISHING/GRINDING   1,475     5,134        0   $2,200          $0
SCREEN PRINTING                314       362       61   $1,380     $84,528
SEMICONDUCTOR PHOTORESIST      282       282        0   $2,520          $0
SEMICONDUCTOR PHOTORESIST
  STRIPPING                    171       233        0   $2,520          $0
SEMICONDUCTOR WAFER CLEANING   343     1,028        0   $2,520          $0
SEMICONDUCTOR - CHEMICAL
  ETCHENTS                     171       233        0   $2,520          $0
SEMICONDUCTOR - DIFFUSION &   ION IMPLANT                   49     7,339        0   $2,520          $0
SEPARATION                      49        49        0   $1,120          $0
SOLDERING                   22,901    96,171    9,730   $1,140 $11,092,750
STAMPING/SHAPING/MOLDING/
  PRESSING                   3,328    17,434    1,607   $2,900  $4,660,122
STERILIZATION                1,480       957        0     $710          $0
WELDING/BRAZING              1,153     2,318        0   $1,140          $0
WELDING/SOLDERING           13,988    62,094    2,024   $1,140  $2,307,300
ZSUBTOTAL                  205,154   765,639   49,499          $89,615,064
ZZMAINTENANCE              123,365    25,313    3,338     $520  $1,735,608
ZZTOTAL                    328,519   790,953   52,837          $91,350,672

SIC 40, 45, 47 - TRANSPORTATION

                         FREQUENCY      TOTAL  COSTED           TOTAL COST
                                OF      WORK     WORK     UNIT         FOR
PROCESS                    PROCESS  STATIONS STATIONS     COST     PROCESS
__________________________________________________________________________

ASSEMBLY                         7       372        0   $1,140          $0
CLEANING                       213       213        0     $710          $0
DEICING                         15        30            $2,410          $0
ENGINE FUELING/OPERATION       272     4,361        7     $560      $4,161
HANDLING SPILLS/LEAKS           58        58        0       $0          $0
LOADING/OFFLOADING/
  RECEIVING/HANDLING           471    39,011        7   $1,120      $8,322
MACHINING/GRINDING/WELDING/
  BRAZING                        7         0        0   $1,140          $0
MAINTENANCE ACTIVITIES         352       401        0       $0          $0
PAINTING/COATING                15         7        0   $3,070          $0
RECEIVING/SHIPPING             242     2,586    2,558   $1,120  $2,864,591
REFUELING                       22       171       22     $560     $12,483
SPECIAL CARE OF LADING
  SERVICES                      36        29        0     $560          $0
WELDING/BRAZING                242       455        0   $1,140          $0
ZSUBTOTAL                    1,953    47,694    2,595           $2,889,559
ZZMAINTENANCE               42,025     1,403      132     $520  $5,723,634
ZZTOTAL                     43,979    49,097    2,727           $8,613,193

SIC 49 - ELECTRIC, GAS, AND SANITARY SERVICE UTILITIES

                         FREQUENCY      TOTAL  COSTED           TOTAL COST
                                OF      WORK     WORK     UNIT         FOR
PROCESS                    PROCESS  STATIONS STATIONS     COST     PROCESS
__________________________________________________________________________

ASSEMBLY                         5         3        0   $1,140          $0
BOILER/FURNACE PREP/OPERATION  630     1,240      368     $710    $262,310
BREAKING UP WASTE               68       136        0  $14,000          $0
CHEMICAL PREPARATION/
  APPLICATION                   72        72        0  $14,000          $0
CLEANING                        14         0        0     $710          $0
COLLECTION/TRANSPORT           524    10,414    6,111     $520  $8,403,203
CONDENSATE COLLECTION          151       917      584     $520  $1,027,855
DETOXIFICATION                   5        15        3   $2,410      $6,034
ENGINE FUELING/OPERATION     1,022     2,124      130     $560     $72,676
HANDLING SPILLS/LEAKS          754     9,906        0        0          $0
INCINERATION                    14        55        0  $14,000          $0
LAB PROCEDURES: TISSUE
  STAINING & FIXING              3         3        0       $0          $0
LOADING/OFFLOADING/
  RECEIVING/HANDLING            14        28        0   $1,120          $0
MAINTENANCE ACTIVITIES       4,781    18,397        0       $0          $0
ODORANT ADDITION               384     1,334    1,170       $0  $2,059,194
OTHER                          167       246        0   $1,120          $0
PAINTING/COATING                88        90        0   $3,070          $0
PRINTING                         3        15        0   $1,380          $0
PROCESS INSP/SUPERVISION/
  QUAL CONTROL                  99       498       68   $1,120     $75,713
PUMP STATION FUELING/ENGINE
  FUMES                         65        65       65     $560     $36,338
RECEIVING/SHIPPING               3         5        0   $1,120          $0
RECYCLING/RECLAMATION           44        47       14  $21,900    $310,010
SAMPLING                        79       230        0   $1,120          $0
SAMPLING OF PIPE LINES         295     1,474    1,474   $1,120  $1,651,346
USE OF CHEMICAL ADDITIVES    1,145     4,382        0   $1,120          $0
USE OF DISINFECTANTS &   SOLVENTS                     228       228        0   $2,410          $0
VENTING                         14     1,380        0       $0          $0
WATER PURIFICATION             540     1,251        0  $14,000          $0
WATER TREATMENT                655     2,167       92  $14,000  $1,281,486
WELDING                        111       212        5   $1,140      $5,708
WELDING/BRAZING                162       239        0   $1,140          $0
WOOD PRESERVATION               65         0        0   $3,070          $0
ZSUBTOTAL                   12,200    50,171   10,083          $15,191,877
ZZMAINTENANCE               15,812    11,056    1,345     $520 $22,817,441
ZZTOTAL                     28,012    61,226   11,428          $38,009,318

SIC 50, 51 - WHOLESALE TRADE

                         FREQUENCY      TOTAL  COSTED           TOTAL COST
                                OF      WORK     WORK     UNIT         FOR
PROCESS                    PROCESS  STATIONS STATIONS     COST     PROCESS
__________________________________________________________________________

BALING/COMPACTING              217       295       60  $14,000    $838,812
BLENDING/MIXING/FORMULATING  2,264     4,048      954   $4,740  $1,844,592
BREAKING UP WASTE               38       109        0  $14,000          $0
CHEMICAL PREPARATION/
  APPLICATION                  684     1,654      182  $14,000    $321,174
CHIPPING GRINDING                8        60        0   $1,140          $0
COATING/SPRAYING/FINISHING/
  LAYUP                        457     1,066      913   $3,070  $2,804,276
COLLECTING                     518     4,687        0     $520          $0
CRUSHING/GRINDING/CALCINING     30        30        0   $4,740          $0
DRILLING/CUTTING/FLAME-JET
  LANCING                        4        11        0   $1,140          $0
DRYING/BAKING                   34       131        0   $4,740          $0
MAINTENANCE ACTIVITIES           4        15        0     $520          $0
METAL MELTING/POURING/CASTING    4        11        0   $1,820          $0
OTHER                           67       116        0   $1,140          $0
PACKAGING/REPACKAGING        3,144     5,071    1,752   $1,760  $3,084,385
PESTICIDE PREPARATION/
  APPLICATION                   30         0        0     $520          $0
POLISHING (SURFACE)/GRINDING    30        30        0   $1,140          $0
PROCESS INSP/SUPERVISION/
  QUAL CONTROL                   4        95        0   $1,120          $0
REACTION/FERMENTATION            8        15        0   $1,120          $0
RECEIVING/SHIPPING           8,725    13,696    6,864   $1,120  $7,470,731
RECYCLING/RECLAMATION            4         4        4  $21,900     $82,794
SAMPLING                       152       152        0   $1,120          $0
SANDING/POLISHING/GRINDING     152       304        0   $2,200          $0
SORTING                        794     1,472       30  $14,000    $419,406
WASHING                        152       152        0       $0          $0
WELDING                          4        11        0   $1,140          $0
ZSUBTOTAL                   17,526    33,237   10,759          $16,866,172
ZZMAINTENANCE               39,371       800       19     $520    $344,858
ZZTOTAL                     56,897    34,037   10,778          $17,211,030

SIC 55, 75 - AUTO DEALERS AND REPAIR

                         FREQUENCY      TOTAL  COSTED           TOTAL COST
                                OF      WORK     WORK     UNIT         FOR
PROCESS                    PROCESS  STATIONS STATIONS     COST     PROCESS
__________________________________________________________________________

ASSEMBLY                         9        66        0   $1,140          $0
CLEANING                    83,012   342,202    2,458     $710  $1,745,425
COATING/SPRAYING/FINISHING/
  LAYUP                          3         0        0   $3,070          $0
CONFINED SPACE EXPOSURE TO
  EXHAUST FUME              57,023   224,767   23,623      $80  $1,889,810
COATING/PAINTING            23,623         0        0   $3,070          $0
FLOAT PROCESS                    3         0        0      $90          $0
MACHINING/GRINDING/WELDING/
  BRAZING                        3        47        0   $1,140          $0
MAINTENANCE ACTIVITIES      49,086   155,359        0      $80          $0
MATERIALS MANUFACTURE/
  FABRICATION                    3         3        0   $1,140          $0
OTHER                       11,811         0        0      $90          $0
OTHER SOLVENT USE           14,592    50,937        0     $710          $0
PAINT STRIPPING             14,795    41,509       96     $710     $68,069
PAINTING/COATING            33,997   127,958        0   $3,070          $0
PUMP STATION FUELING/
  ENGINE FUMES              23,623    94,491        0     $560          $0
REACTION/FERMENTATION       11,811         0        0   $2,900          $0
RECEIVING/SHIPPING              70       211        0   $1,120          $0
REFRIGERANT CHARGING        11,811         0        0   $3,070          $0
STAMPING/SHAPING/MOLDING/
  PRESSING                       3        22        0   $2,900          $0
TORCH CUTTING               11,811         0        0   $2,520          $0
WELDING                     45,594   164,157    7,081   $1,140  $8,072,535
ZSUBTOTAL                  392,685 1,201,729   33,258          $11,775,839
ZZMAINTENANCE              257,267    76,073    3,400     $520  $7,918,216
ZZTOTAL                    649,952 1,277,801   36,658          $19,694,055

SIC 76 - MISCELLANEOUS REPAIR SERVICE

                         FREQUENCY      TOTAL  COSTED           TOTAL COST
                                OF      WORK     WORK     UNIT         FOR
PROCESS                    PROCESS  STATIONS STATIONS     COST     PROCESS
__________________________________________________________________________

CLEANING                         2         4        0     $710          $0
GLUEING/HOT PRESSING         3,982     4,837       16   $3,070     $48,168
OTHER                           13        13        0   $2,200          $0
PAINT STRIPPING              2,348     2,820        0     $710          $0
PAINTING/COATING             2,381     2,492        0   $3,070          $0
SANDING/POLISHING/GRINDING   3,876     6,151        0   $2,200          $0
WELDING/BRAZING              4,052     7,830    2,120   $1,140  $2,417,367
ZSUBTOTAL                   16,653    24,145    2,836           $2,465,536
ZZMAINTENANCE               15,095     4,260      509     $520    $344,257
ZZTOTAL                     31,747    28,405    3,345           $2,809,793
__________________________________________________________________________

References

1. Dun and Bradstreet, Inc., 1985 Count of Establishments. (Database)

H. Economic Impact, Regulatory Flexibility Analysis and Environmental Impact Assessment

Economic Impact

The economic impacts discussed in this chapter have been estimated following an analysis of data collected through a nationwide sample survey of about 5,700 establishments. Two alternative polar assumptions were used in this analysis.

* Perfectly Elastic Demand or Zero Cost-Passthrough: All compliance costs are absorbed by the firm in the form of reduced profits. This assumption is the "worst case" scenario, where the maximum reduction in pre-tax profits to the firm (and industry) results.

* Perfectly Inelastic Demand or Total Cost-Passthrough: All compliance costs are passed on to the consumer sector in the form of higher prices. From the perspective of the firm, this is the "best case" scenario. The resulting price increase would be the maximum theoretically possible.

Two points should be noted. First, for the majority of industry sectors, neither assumed market structure would be accurate. In practice, the impacts will almost always produce a price increase smaller than the inelastic demand projection and a reduction in profits smaller than that predicted under perfectly elastic demand conditions. Second, increased firm productivity would mitigate any adverse economic effects of the final standard. Productivity effects would be related to reduced worker illness, absence and turnover. In addition, knowledge of improved workplace health conditions could result in higher workforce morale and productivity. The firm would enjoy lower employee training costs (due to the reduced turnover rate) and lower medical benefit and worker compensation claims. Overall productivity increases would be realized by firms that use a relatively fixed-factor production process (i.e., low elasticities of substitution between labor and other factors of production). It is difficult to estimate the magnitude of these productivity and cost reducing effects. Any estimated economic costs of compliance would have to be adjusted downward to reflect these effects. Since data were not available to make any offset estimates, the economic effects of the standard identified in this chapter are overstated.

In addition, OSHA is allowing a phase-in period, up to five years, for engineering controls. Respirator use will be allowed during this period. Economic costs presented in this chapter will be overstated to the extent that capital expenditures are delayed during the phase-in period.

For this analysis, OSHA used a percentage reduction in profits approach to obtain estimates of the short-run economic impacts under the assumption of perfect demand elasticity. These estimates were obtained by using the following formula:


    Percentage Reduction = New Profits - Old Profits
                           _________________________
                                   Old Profits

where:
New Profits = (Old Pre-tax Profits - Compliance Costs) x
    (1 - Old Profits) = (return on Sales)*(Total Sales).

These calculations were performed at the two-digit SIC level for firms in large and small size-class stratifications (above and below 20 employees). The data used to obtain these estimates was based on Dun and Bradstreet company files [1;2].

The potential impact on prices was used to estimate the market consequences under the second assumption of inelastic demand. Total sales values for 1985 were used, the year for which the compliance costs were estimated. (Total sales represent the totality of production that leaves the establishment, whether it is sold to customers or sent to a parent company in a captive transaction. For industries in the service and trade sectors, total sales data were also used. The rate of return percentage for each industry sector corrected and transformed gross sales data into more accurate and relevant industry profit estimates.) For a given firm-size class, the potential price increase was estimated by dividing the total estimated compliance costs for a firm by the sales of that firm. These estimated price effects were then compared to recent industry price series. The intent of this comparison was to evaluate the impact of the compliance cost-generated price increase in light of recent industry price increase experience.

In this scenario, the potential for international trade implications of the standard was explored. It is anticipated that any international trade effects will not be significant given the small value of domestically produced goods and services which are exported (about seven percent of GDP). Also, the U. S. dollar has recently experienced a sharp decline in value relative to the yen and European currencies. Between February 1985 and December 1987, the trade-weighted value of the U.S. dollar fell 46 percent [3]. This depreciation will likely overwhelm any potential adverse international economic effect of the standard.

In Table H-1 and H-2, the estimated domestic economic impacts are reported for the two polar methodologies. To derive the percentage change in profits and the costs as a percent of sales, industry sales and rate of return (R.o.R.) on sales data were obtained from Dun and Bradstreet. The total sales data are best estimates for industry sectors potentially impacted by the rule.


                          TABLE H-1

           ECONOMIC EFFECTS: NO-COST PASSTHROUGH SCENARIO(1)
_________________________________________________________________________
                                      R.o.R                            %
                   Annual      Total     on  Pre-Reg   Post-Reg    Change
                 Costs(2)   Sales(3)  Sales  Profits    Profits        in
SIC  Industry     ($ mil)    ($ mil) (%)(4)  ($ mil)    ($ mil)   Profits
_________________________________________________________________________
20 FOOD PROD.      33.49  353,780.38   1.9  8,008.04   7,986.29   -0.2715
21 TOBACCO          0.02   74,030.13   5.3  3,923.60   3,923.59   -0.0003
22 TEXT. MILL      29.48   60,735.22   2.7  1,765.42   1,747.59   -1.0100
23 APPAREL PROD.   31.74   74,474.65   2.8  1,813.22   1,793.56   -1.0845
24 LUMBER & WOOD   56.72   57,994.48   3.9  1,974.51   1,931.92   -2.1574
25 FURNITURE       21.08   37,648.27   3.5  1,411.02   1,398.82   -0.8645
26 PAPER PROD.     31.00  103,694.14   3.7  3,778.20   3,761.12   -0.4519
27 PRINTING & PUB. 33.75  134,830.21   4.8  6,471.85   6,444.77   -0.4185
28 CHEMICAL PROD.  35.45  272,759.67   3.7 11,738.80  11,717.79   -0.1790
29 PETRO. REFINING 23.69  196,400.57   2.7  4,964.85   4,952.04   -0.2579
30 RUBBER &      PLASTICS     111.09   86,538.58   4.3  3,423.75   3,343.76   -2.3361
31 LEATHER PROD.    2.41   15,449.56   2.6    401.69     400.03   -0.4127
32 STONE & CLAY    22.46   46,094.04   4.1  1,954.99   1,940.97   -0.7170
33 PRIMARY METALS  70.96  112,564.26   3.3  3,714.62   3,674.83   -1.0712
34 FAB. METALS     39.42  150,146.41   4.0  6,005.86   5,981.33   -0.4084
35 MACHINERY       45.21  345,144.89   5.1  17,602.39  17,573.57  -0.1637
36 ELEC. MACH.     20.67  245,982.70   5.0  12,299.14  12,286.86  -0.0998
37 TRANS. EQUIP.   49.79  365,427.20   3.9  14,520.25  14,485.24  -0.2411
38 INSTRUMENTS      9.63   83,359.57   4.9   3,373.26   3,367.32  -0.1763
39 MISC. MANUF.    15.84   41,870.30   4.4   1,788.56   1,778.14  -0.5825
40 R.R. TRANS.      1.08   43,869.14  10.0   3,969.62   3,969.04  -0.0147
45 AIR TRANS.       3.74  109,538.08   3.6   3,251.40   3,249.38  -0.0621
47 TRANS. SERVICES  3.79   12,254.96   2.7     582.18     580.13  -0.3515
49 ELEC., GAS &      SAN.          38.01  300,254.83   7.0  21,017.84  20,994.71  -0.1100
50 WHOLESALE
     TRADE(5)       3.00   13,853.52   2.0     277.07     274.56  -0.9048
51 WHOLESALE,
     NON-DUR       14.22  113,848.20   1.5   1,726.26   1,718.59  -0.4447
55 AUTO DEALERS    13.55  341,574.50   1.9   6,489.69   6,480.69  -0.1422
72 PERSONAL SERV.  10.87   24,270.74   7.3   1,771.76   1,763.60  -0.4606
73 BUSINESS SERV.   2.42   22,165.94   6.6   1,462.94   1,460.94  -0.1375
75 AUTO REPAIR      6.14   45,750.92   5.1   2,492.19   2,488.29  -0.1563
76 MISC. REPAIR
     SERV.          2.81    2,665.52   5.5     146.60     144.36  -1.5298
80 HEALTH SERVICES  4.44  170,234.25   4.5   7,807.72   7,804.54  -0.0406
_________________________________________________________________________
Source:  U.S. Department of Labor, Occupational Safety and Health
         Administration, Office of Regulatory Analysis.

  Footnote(1) All values in 1985 dollars.
  Footnote(2) Reproduced from Table VI-1.
  Footnote(3) Dun and Bradstreet, Dun's Marketing Identifiers (DMI)
Database.
  Footnote(4). Rate of Return on Sales, Dun and Bradstreet, Industry Norms
Database.
  Footnote(5). Consists of SIC 5093 (scrap and waste materials) only.

                             TABLE H-2

                ECONOMIC EFFECTS: TOTAL-COST PASSTHROUGH
___________________________________________________________________
                                                         Costs as a
                         Annual Costs     Total Sales    Percent of
SIC    Industry          ($ millions)     ($ millions)        Sales
-------------------------------------------------------------------

20     FOOD PROD.               33.49       353,780.38       0.0095
21     TOBACCO                   0.02        74,030.13       0.0000
22     TEXT. MILL               29.48        60,735.22       0.0485
23     APPAREL PROD.            31.74        74,474.65       0.0426
24     LUMBER & WOOD            56.63        57,994.48       0.0978
25     FURNITURE                26.28        37,648.28       0.0560
26     PAPER PROD.              33.00       103,694.14       0.0299
27     PRINTING & PUB           34.39       134,830.21       0.0250
28     CHEMICAL PROD.           38.87       272,759.67       0.0130
29     PETRO. REFINING          23.91       196,400.57       0.0121
30     RUBBER & PLASTICS       121.93        86,538.58       0.1284
31     LEATHER PRODUCTS          2.66        15,449.56       0.0156
32     STONE & CLAY             25.83        46,094.04       0.0487
33     PRIM. METALS             78.24       112,564.26       0.0630
34     FAB. METALS              53.51       150,146.41       0.0263
35     MACHINERY                50.00       345,144.89       0.0131
36     ELEC. MACH.              23.30       245,982.70       0.0084
37     TRANS. EQUIP.            49.79       365,427.20       0.0136
38     INSTRUMENTS              10.75        83,359.57       0.0116
39     MISC. MANUF.             17.29        41,870.30       0.0378
40     R.R. TRANS.               1.09        43,869.14       0.0025
45     AIR TRANS.                3.76       109,538.08       0.0034
47     TRANS. SERVICES           3.81        12,254.96       0.0309
49     ELEC., GAS & SAN.        37.83       300,254.83       0.0127
50     WHOLESALE TRADE(1)        3.13        13,853.52       0.0216
51     WHOLESALE, NON-DUR.      14.80       113,848.20       0.0125
55     AUTO DEALERS             22.72       341,574.50       0.0040
72     PERSONAL SERVICES        10.87        24,270.74       0.0448
73     BUSINESS SERVICES         2.42        22,165.94       0.0109
75     AUTO REPAIRS             10.25        45,750.92       0.0134
76     MISC. REPAIR SERV.        4.86         2,665.52       0.1054
80     HEALTH SERVICES           4.44       170,234.25       0.0026
___________________________________________________________________
Source:  U.S. Department of Labor, Occupational Safety and Health
Administration, Office of Regulatory Analysis.

  Footnote(1). Consists of SIC 5093 (scrap and waste materials) only.

Dun and Bradstreet provided OSHA with this information. The R.o.R. on sales were obtained from summary statistics found in the Dun and Bradstreet Industry Norms Database.

Industry Effects. The estimated economic impact of the rule for firms potentially affected is summarized in Table H-1. These estimates represent the maximum industry impact within a market scenario where none of the costs can be passed onto consumers, and there is no productivity offset to costs.

Data in Table H-1 indicate that the rule will not have a significant impact on profits in most industry sectors. The estimated average change in profits is less than one percent; this amount of profit reduction should not represent a significant economic burden.

The most adversely affected industry sector is SIC 30, Rubber and Plastics, with an estimated 2.3 percent reduction in profits. The only other industry with an impact greater than 2 percent is SIC 24, Lumber and Wood (2.2 percent). However, even in the worst case, OSHA believes the standard is economically feasible. In reality, the reduction in profits will be less because part of the costs will be passed on to consumers, and because profitability in these industries since 1985 (the year for which the cost impact was estimated) has improved [9].

Consumer effects were estimated using a "full cost passthrough"

scenario. As demonstrated by the estimates summarized in Table H-2, the impacts on market prices will not be significant. No price increase would exceed one half of one percent. Changes of this magnitude are within general price movements recorded by producer price and other price indices.

During the public hearing and comment period OSHA received comments concerning the economic impact of the standard. Comments were primarily concerned with the following industries and substances.

SIC 20: The meat packing and processing industries (SICs 2011 and 2013) expressed concern over the impact a carbon disulfide PEL of 1 ppm would have on the cellulosic casings industry (SIC 3079). Industry representatives in all three of these SICs believe that if OSHA sets a PEL which is economically difficult to meet, domestic production of the meat casings will dramatically decrease [Ex. 3-659; 3-756; 3-757; 3-898]*. Although foreign markets already supply casings to the U.S. meat industry, this industry believes that foreign markets will not be able to absorb a dramatic change in demand for casings [Ex. 3-898; 3-897; 3-756].


______________
  Footnote(*) All Exhibit [Ex.] numbers refer to the material in Docket
H-020, the official record of this rulemaking.  References to the
transcripts of the public hearings, available in the docket, are
identified as "Tr." followed by the date of the hearing and the page
numbers of the transcript.

OSHA has decided to increase the PEL for carbon disulfide from the proposed limit of 1 ppm to a standard of 4 ppm. Compliance with a 4 ppm PEL will be easier for casing manufacturers. OSHA believes that domestic production will be unaffected by a carbon disulfide PEL of 4 ppm, particularly since respirators may be used in difficult to control operations.

SIC 23: See SIC 72. SICs 24 and 25: The wood and furniture industries argued that to comply with the proposed PEL of 5 mg/m(3) for softwood and 1 mg/m(3) for hardwood, the majority of establishments will have to engineer to 1 mg/m(3) for all wood dust since most establishments process both soft and hard wood. Most commenters concluded that they, as well as most members of their industry, would go out of business if they had to control wood dust exposures to 1 mg/m(3).

Several industry organizations submitted comments and information concerning the potential economic impact of these PELs on their industry [Ex. 3-626; 3-627; 3-748; 3-899; 3-951; 3-627]. The U.S. Small Business Administration (SBA) believes "it is likely that the wood dust PELs of 1 mg/m(3) for hard wood dust and 5 mg/m(3) for soft wood dust are not economically feasible for small facilities with fewer than 20 employees," and that a "5 mg/m(3) standard...may be more economically feasible for affected industries" [Ex. 3-951]. This recommendation was supported by many comments from industry [Ex. 3-627; 3-768; 3-750; 3-917].

OSHA's final rule establishes a 5 mg/m(3) PEL for all wood dust except Western red cedar, for which a 2.5 mg/m(3) limit is established. OSHA believes that the 2.5 mg/m(3) standard for this allergenic wood is feasible since 90% of the firms using Western red cedar are located in the state of Washington which has already adopted a 2.5 mg/m(3) PEL.

OSHA's estimated costs for compliance with the final PELs are significantly lower than those corresponding to the proposed PELs of 1 and 5 mg/m(3). The economic impact of the standard on these wood processing industries reflects this decrease in cost. These reduced costs amount to 2% of profits. (The actual effects on profits will be even less since some costs will be passed on to consumers.) Industries in SIC 24 had strong domestic and foreign markets in 1987.

Price increases averaged 5% over 1986. As a result of increasing prices, demand, and output, profits increased for firms in this industry. The 1988 U.S. Industrial Outlook predicted a 4.1% growth (in constant dollars) of shipments of wood products in 1988. These recent developments indicate that the economic impact of complying with a 5 mg/m(3) PEL for wood dust will be less than the impact presented in Table H-1 [9].

In addition, furniture product shipments in SIC 251 increased 4.5% in 1987 (constant dollars) [9]. This will make it easier for firms in this SIC to absorb the costs imposed by this rule, and the economic impact will actually be less than that estimated by OSHA.

SIC 30: Establishments in SIC 308 submitted comments and information concerning OSHA's proposed standard of 50 ppm for styrene. The Styrene Information Research Council (SIRC) submitted cost and impact estimates of the proposed PEL on SIC 30 and SIC 37 [Ex. 3-742]. OSHA has examined this information and, based on reasons outlined in the cost chapter of this document (Chpt. G), has determined that the survey performed for this rulemaking provides the most accurate and up-to-date information on employee exposures and cost of controls. The costs OSHA estimates are less than .44% of profits and are economically feasible. In addition, the value of shipments of rubber and plastics products have been increasing since 1985 (for which costs and impacts were estimated), and growth in shipments is projected to be 2% in 1988 [9]. Increased profitability should offset the economic consequences of compliance.

Comments were received from the manufacturers of cellulosic casings (SIC 3079) concerning OSHA's proposed PEL of 1 ppm for carbon disulfide. There are three companies which currently produce these casings. A study of engineering controls required to comply with this PEL, as well as the costs of these controls, was submitted by industry [Ex. 8-45]. This study indicated high costs to control to 1 ppm. OSHA has subsequently adopted a 4 ppm PEL for the final rule. OSHA concludes that industry can comply with this level. (According to docket evidence, at least one company already operates at 10 ppm or less [Ex. 3-945].) The costs are clearly feasible.

SIC 33: Representatives of the American Iron and Steel Institute indicated in their comments that because of the "fragile financial condition" of the industry, capital investment for equipment such as engineering controls has been limited [Ex. 72, pg. 32]. Representatives expressed concern over the costs to comply with proposed OSHA PELs for several different substances and the economic impact on competition with foreign steel producers [Ex. 72, pg. 3, and pp.33-34]. Although the dollar has recently depreciated in value relative to the yen and European currencies, depreciation of the dollar relative to the value of currencies of steel-producing countries has been gradual [9]. It is likely that the dollar's depreciation will not be as beneficial to the steel industry as it will for other industries. However, exports of steel-intensive products (excluding motor vehicles) has increased due to the dollar's depreciation (by June of 1987, the volume of exports was 24% over the 1986 level) [9]. Although the steel industry is not growing rapidly, it is certainly not experiencing the downturn of the early 1980's, and the impact of compliance costs should not be as detrimental as the industry predicted [Ex. 72].

In addition, the PELs for iron oxide and aluminum metal dust, two substances which constituted a significant part of the costs estimated for this industry in the 1987 proposal, will not be changed from the OSHA standard currently in effect. OSHA recognizes the special feasibility problems of complying with the proposed PELs for hazardous substance exposures in the steel industry, and is allowing the use of respirators in operations where carbon monoxide and sodium dioxide are present. These changes will substantially decrease costs to the industry, and hence will lessen the economic impact.

SIC 37: Styrene exposures in the manual layup/sprayup operations in the boat building industry are difficult to control through engineering methods due to the nature of the operation and the small space within which the styrene is applied [Ex. 3-742]. Evidence submitted to the docket suggests typical exposures in this industry are below 50 ppm except in the layup/sprayup operation [Ex. 3-742]. OSHA is permitting respirator use in these operations in this industry in view of special compliance problems. The costs are low (.24%) in relation to profits. OSHA concludes there will be no adverse economic impact on the industry.

SIC 51: Concern was expressed by the National Grain and Feed Association (NGFA) on behalf of the grain elevator operators/grain handlers in SIC 5153 and the National Cotton Council regarding the feasibility of the proposed PEL of 4 mg/m(3) for grain dust [Ex. 3-752 and Ex. 3-1080]. For the final rule, OSHA has established a PEL of 10 mg/m(3). Most employee exposures are at or below 10 mg/m(3), (see Chapt. G, pg. 19).

OSHA's assessment of the economic impact of the proposal by two-digit SIC was criticized as being too general an approach for estimating the economic consequences of the rule on industry subsectors [Ex. 3-752]. The economic impact of the standard is based on the costs presented above in Chapter G. These costs are based on an industry survey conducted by OSHA for this rulemaking which gathered data at the four digit level. However, the survey was designed to be statistically meaningful at the cell level (two or three digit SIC level). There would be more uncertainty at the four digit level. Much four digit data were in the record and OSHA developed more when requested by participants. As discussed above, OSHA concludes that its cost estimate for SIC 51 (which includes many grain elevators) is accurate. The costs demonstrate economic feasibility even if all costs were borne by SIC 5153 (grain elevators).

SIC 72: In the proposed standard, OSHA indicated an intention to change the PEL for perchloroethylene to 50 ppm. Employees are exposed to this chemical during a wet-to-dry industrial process used in the dry cleaning, laundry, and garment sector (SIC 721). Comments received from the International Fabricare Institute indicated that by 1992, almost all machines used by dry cleaners will be dry-to-dry, a process which has reduced exposures to perchloroethylene [Ex. 3-671]. OSHA believes that industry can comply with a lower PEL of 25 ppm within the four year phase-in period through the normal course of capital replacement as dry-to-dry process equipment is substituted for wet-to-dry process equipment.

OSHA is sympathetic to the circumstances of the number of small businesses in this SIC. OSHA has stated in the Preamble discussion that a phase-in period, up to five years, will be allowed for engineering controls. If it appears that there will be a significant economic difficulty for small dry cleaning operations to convert to new equipment or to retrofit within the time period permitted by the Standard, OSHA will consider extending the phase-in period for firms in this industry.

Regulatory Flexibility Analysis

In accordance with the Regulatory Flexibility Act (P.L. 96-353, 94 Stat. 1164 [5 U.S.C. 601 et seq.]), OSHA has assessed the impact of the rulemaking on large and small establishments. For this assessment, large establishments are defined as those with 20 or more employees and small establishments as those with 19 or fewer employees. The results of this assessment are summarized in Table H-3.


                              TABLE H-3

              ECONOMIC IMPACTS BY ESTABLISHMENT SIZE
________________________________________________________________
                                  Percentage Change in Profits
                                  ____________________________
SIC      Industry                 Large                  Small
________________________________________________________________

20       FOOD PROD.               - 0.1526             - 3.0770
21       TOBACCO                  - 0.0003               0.0000
22       TEXT. MILL               - 0.7442             - 7.0804
23       APPAREL PROD.            - 0.7916             - 3.4067
24       LUMBER & WOOD            - 0.5895             -10.4061
25       FURNITURE                - 0.7701             - 1.8987
26       PAPER PRODUCTS           - 0.4440             - 0.6384
27       PRINTING & PUB.          - 0.0536             - 1.9402
28       CHEMICAL PROD.           - 0.1390             - 0.9853
29       PETRO. REFINING          - 0.2619             - 0.0458
30       RUBBER & PLASTICS        - 0.7788             -28.5020
31       LEATHER PROD.            - 0.1838             - 3.4971
32       STONE & CLAY             - 0.5564             - 1.2857
33       PRIMARY METALS           - 1.0084             - 2.1953
34       FAB. METALS              - 0.2889             - 1.5027
35       MACHINERY                - 0.0988             - 1.8448
36       ELEC. MACH.              - 0.0766             - 1.1400
37       TRANS. EQUIP.            - 0.0883             -21.2970
38       INSTRUMENTS              - 0.1201             - 1.6686
39       MISC. MANUF.             - 0.3489             - 1.9125
40       R.R. TRANS.                n/a                  n/a
45       AIR TRANS.                 n/a                  n/a
47       TRANS. SERVICES            n/a                  n/a
49       ELEC., GAS & SAN.          n/a                  n/a
50       WHOLESALE, TRADE(1)      - 0.6786             - 1.2044
51       WHOLESALE, NON-DUR       - 0.1638             - 0.7812
55       AUTO DEALERS             - 0.0948             - 0.2293
72       PERSONAL SERV.           - 0.2395             - 0.6369
73       BUSINESS SERV.           - 0.1374             - 0.1376
75       AUTO REPAIR              - 0.4836             - 0.0786
76       MISC. REPAIR SERV.       - 0.6105             - 1.8934
80       HEALTH SERVICES          - 0.0736             - 0.0295
________________________________________________________________
Source:  U.S. Department of Labor, Occupational Safety and Health
Administration, Office of Regulatory Analysis.

  Footnote(1). Consists of SIC 5093 (scrap and waste materials) only.

Industry sales and profit estimates were based on data from Dun and Bradstreet and the Department of Commerce 1982 Census of Manufactures [5], Wholesalers [6], Retailers [7] and Services [8]. Sales and profit data for selected transportation sector industries (SIC 40, 45, 47 and 49) were not available for use in this assessment.

The information summarized in Table H-3 indicates that with three exceptions small firms will not have important adverse impacts.

Data for small establishments in SIC 24 (Lumber and Wood), SIC 30 (Rubber and Plastics), and SIC 37 (Transportation Equipment Manufacturers), show the potential for more significant changes in profits. In the case of SIC 24, many small businessmen and their representatives testified and supported the final standard. This suggests that the impact will be manageable.

It should be noted that these negative effects result in part from the extreme assumption of perfectly elastic demand. An important ameliorating factor for each firm will be its ability to pass through additional costs to the consumer. The ability of individual firms to do this will be dependent upon product demand elasticities. It is expected that most impacted firms will be able to pass through some portion of their increased costs.

Environmental Impact Assessment

This assessment has been prepared in accordance with provisions of the National Environmental Policy Act (NEPA) (42 U.S.C. 4325 et seq.) as well as the regulations of the Council on Environmental Quality (40 CFR Part 1500), and DOL-NEPA Compliance Procedures (29 CFR Part 11).

OSHA has reviewed the standard and the information contained in the secondary data bases, as well as the information submitted by the contractors' industry experts and submissions by the public to the record during the course of this rulemaking, and has concluded that no significant environmental impacts are likely to occur as a result of this action.

Two environments may be affected by an OSHA regulatory action: (1) the workplace environment; and (2) the general human environment external to the workplace, including impacts on air and water pollution, solid waste, energy, and land use. Usually OSHA regulations have their most significant impacts on the workplace environment since this environment is under the Agency's jurisdiction. Lower and new PELs will benefit the workplace environment because they will reduce worker exposure to toxic substances.

In most cases, the effects of previous OSHA regulations on the external environment have been negligible because of their limited scope and application. Similarly, there is no evidence to indicate that there would be any significant adverse impacts to the external environment as a result of this standard. As with other OSHA regulations in the past, however, there may be a potential benefit to the environment. The potential benefits and other impacts are briefly summarized here.

Air Pollution. Because of the nature of the emission standards of the Environmental Protection Agency (EPA) (40 CFR Part 61), many industry operations already use engineering controls to reduce the amount of emissions to the atmosphere. This practice is not expected to change as a result of the rule. OSHA anticipates that controls already in place will continue to operate effectively in reducing emissions under the revised standard. Thirteen of the chemicals addressed in this standard have been recognized by EPA as air pollutants. These are listed below;

* Beryllium * Carbon Monoxide * Epichlorohydrin * Ethylene dichloride * Hexachlorocyclopentadiene * Mercury * Methyl chloroform * Nitrogen dioxide * Ozone * Perchloroethylene * Sulfur dioxide * Toluene * Trichloroethylene

Water Pollution. EPA regulates over 100 of the chemicals addressed in this standard under the Clean Water Act of 1977 (33 USC 1251 et seq.). EPA's effluent limitation guidelines (40 CFR Part 427) include (1) standards of performance for all new point sources within specified categories and (2) pretreatment standards for new plants discharging to municipal sewer systems. These limitations would serve to prevent the discharge of effluents into the environment without prior treatment. Moreover, the Federal Water Pollution Control Act Amendments of 1972 required that wastewater effluents be treated by the best practicable technology (BPT) by December 31, 1977 and that the best available technology (BAT) economically achievable be used by December 31, 1983. The EPA effluent limitations establish the degree of effluent quality necessary to meet the BPT and BAT requirements. The BAT and pretreatment standards would essentially mean no discharge of process wastewater to navigable waters and no discharge of incompatible pollutants. These requirements will not change as the result of this proposal and where they continue to be met, effluent quality will not be altered.

Solid Waste Disposal. It does not appear that there would be any significant change in present waste disposal practices for over 80 chemicals addressed by this rule, or in the maintenance of waste disposal sites. EPA's national emissions standards will continue to provide for the control and maintenance of active and inactive disposal sites and require no visible emissions from these sites.

Energy And Land Use. The implementation of required engineering controls could result in an increase in total energy requirements or costs for general industry. This would be particularly true where controls are not in place. Where general exhaust ventilation is used, there is the expense of heating or cooling the replacement air brought in from the outside. These costs, plus the cost of vacuuming, where necessary, have been included in the annual costs estimated in Chapter G. In terms of land use, OSHA does not project any significant impact on land use plans, policies or controls. OSHA does not anticipate any significant impact on the short term uses of man's environment or upon the maintenance of long-term productivity.

Other Impacts. The standard could also have other impacts that may affect the external environment. The standard could encourage the further use, research, and development of suitable substitutes for hazardous chemicals. This, in turn, would result in a positive environmental effect because fewer hazardous chemicals would be used, emitted to the air, discharged as wastewater effluent, or disposed of as solid waste. The magnitude or probability of these impacts, however, is impossible to quantify.

Overall, the projected impacts of the standard on the external environment are not expected to be significant in view of EPA's regulation of air emissions, water effluents, and solid waste disposal methods.

Summary

Based on the data summarized in Tables H-1 and H-2 and historical information, and information submitted by the public during this rulemaking procedure, OSHA has concluded that the economic impacts of the standard are clearly feasible for industry sectors and subsectors. However, the estimates indicate that some small establishments in SICs 24, 30, and 37 may experience a greater impact than larger entities. The rule is not expected to have an adverse effect on the environment.

References

1. Dun and Bradstreet, Inc. Industry Norms and Key Business Ratios. 1985. (Database) 2. Dun and Bradstreet, Inc. Dun's Market Identifiers. 1987. (Database) 3. Federal Reserve Board of Governors. Economic Trends. Cleveland:

Federal Reserve Bank of Cleveland, December 1987.

4. Liebowitz, S.J. "What Do Census Price-Cost Margins Measure?" Journal of Law and Economics 25:231-46, 1982.

5. U.S. Department of Commerce, Bureau of the Census. 1982 Census of Manufactures, Industry Series (selected series). Washington D.C.: U.S. Government Printing Office.

6. U.S. Department of Commerce, Bureau of the Census. 1982 Census of Wholesale Trade, Establishment and Firm Size (Including Legal Form of Organization). Washington D.C.: U.S. Government Printing Office.

7. U.S. Department of Commerce, Bureau of the Census. 1982 Census of Retail Trade, Establishment and Firm Size (Including Legal Form of Organization). Washington D.C.: U.S. Government Printing Office.

8. U.S. Department of Commerce, Bureau of the Census. 1982 Census of Service Industries, Establishment and Firm Size (Including Legal Form of Organization). Washington D.C.: U.S. Government Printing Office.

9. U.S. Department of Commerce, International Trade Administration. 1988 U.S. Industrial Outlook. Washington, D.C.

Supplement 1 - Technical Description of the Sample Survey

1. Introduction

This appendix contains a description of the statistical methodology employed to design and implement the PEL survey. The following topics will be discussed:

* Survey objectives;
* Sampling frame selection;
* Stratification;
* Sample size determination;
* Estimation procedures;
* Data collection method;
* Variance estimation;
* Treatment of non-sampling errors; and
* Survey instrument.

2. Survey Objectives

Surveys are frequently designed to produce a set of estimates at a predefined level of accuracy. this requires defining the set of quantities to be estimated and specifying their levels of accuracy. Since many variables may ultimately be estimated from the survey, and since no single design can be optimal for all estimates simultaneously, it is customary to define the most important variables for estimation. For this survey, the following variables were identified as those motivating the survey design:

* Cost to industry of the proposed set of new permissible exposure limits (as a group);

* Number of workers potentially exposed to toxic substances; and
* Number of workers affected by the proposed regulations.

Statistical theory dictates that responses be concentrated both in groups which have the highest variability with respect to these variables and in groups representing the majority of establishments in the population. No hard information relating to the variability of the variables mentioned above by industry sector or other relevant breakdown was available at the outset of the survey. Hence, the variability in the number of employees was used as a variability measure. Consistent with the notion that the variability of numbers exposed as well as the variability of cost required to remedy an overexposure are highest in the largest companies, the sample was designed to include a higher proportion of larger establishments.

The sample was drawn so as to insure that the relative standard error (RSE) estimates (the ratio of the sample standard error to the mean) was within predetermined bounds. The relative standard error is a measure of the accuracy of each estimate. A relative standard error of 5 percent means that the standard error of the estimate is equal to 5 percent of that estimate. This can be interpreted as saying that the estimate is within two standard errors or 10 percent of the true value with 95 percent probability. Since risks were judged to be different in different sectors, OSHA selected a 5 percent relative standard error in the industries using the most chemicals, 7.5 percent in industries with moderate use of chemicals and 10 percent in the service sectors. A table of design specifications is included in Section A.5 below.

3. Sampling Frame Selection

The Dun and Bradstreet (D&B) listing was chosen for the PEL survey sampling frame (a listing of establishments from which sample units are selected). This is a nationally based list, containing establishment names as well as each establishment's address, telephone number, SIC code, and number of employees. The Dun and Bradstreet database is regularly refined (every six months) thus minimizing the probability of obtaining out of business or out of scope (e.g., wrong SIC code) establishments when using the frame. The D & B is a commercial listing and its use does not violate any confidentiality requirement associated with other frames available to particular agencies in the government.

4. Stratification

Thirty-four groupings of industries (estimation cells) were chosen to be examined for the PEL study. The cell definitions were determined by grouping together industry sectors defined by Standard Industrial Classifications (SICs) which share similar processes and procedures. The cell definitions used for the PEL survey are given in TABLE 1-1.


                           TABLE 1-1

                  Definitions of Estimation Cells

Cell Number   SIC Codes Included     Description
___________   __________________    _____________________________________
     1               243             Millwork, Veneer, Plywood
     2               245             Wood Buildings & Mobile Homes
     3               249             Misc. Wood Products
     4               25              Furniture
     5               26              Paper Products
     6               27              Printing & Publishing
     7               281             Indust. Inorganic Chems
     8               282             Plastics & Syn. Resins
     9               283             Drugs
    10               284             Soaps, Detergents & Cleaning
    11               285             Paints, Varnishes, Lacquers
    12               286             Indust. Organic Chems
    13               287             Agricultural Chemicals
    14               289             Misc. Chemical Products
    15               291             Petroleum Refining
    16               295             Paving & Roofing Materials
    17               299             Misc. Petroleum Products
    18               308             Misc. Plastic Products
    19               30 (not 308)    Plastics & Rubber
    20               311             Leather Tanning
    21               31 (not 311)    Leather & Leather Products
    22               32              Stone & Clay
    23               33              Primary Metals
    24               34              Fabricated Metals
    25               35              Machinery
                     36              Electrical Machinery
                     38              Instruments
                     39              Misc. Manufacturing
    26               40              R. R. Transportation
                     44              Water Transportation
                     45              Air Transportation
                     47              Transportation Services
    27               46              Pipelines
    28               49              Electrical, Gas & Sanitary
    30               5093            Misc. Durable Goods
                     5153            Grain
                     5161            Chemicals & Allied Products
                     5191            Misc. Farm Supplies
                     5198            Misc. Paints, Varnishes
    31               55              Auto Dealers
                     75              Auto Repair
    32               7211            Power Laundries, Family & Commercial
                     7213            Linen Supply
                     7215            Coin-operated Laundries & Cleaning
                     7216            Drycleaning Plants, except Rug
                     7218            Industrial Launderers
                     7219            Laundry & Garment Services, nec
                     7221            Photographic Studios, Portrait
                     7231            Beauty Shops
                     7241            Barber Shops
                     7251            Shoe Repair & Shoeshine Parlors
                     7261            Funeral Service & Crematories
                     7299            Miscellaneous Services, nec
                     7332            Blueprinting & Photocopying Services
                     7342            Disinfecting & Pest Control Services
                     7395            Photofinishing Laboratories
    33               7641            Furniture Repair
                     7692            Welding Repair
    34               80              Heath Services
    99               37              Transportation Equipment
_________________________________________________________________________

For each estimation cell, units on the Dun and Bradstreet sampling frame
were classified into one of the four size classes listed below:

                      Number of
Size:                 employees
  1....................1 to 19
  2....................20 to 99
  3....................100 to 249
  4....................250 and above

For each size class stratum within a cell, the establishments on the frame were further identified by their four digit SIC classification (within the two or three digit sample cell). A separate systematic sample was then selected in each estimation cell/size class stratum. This procedure was accomplished by first selecting one case at random in the size class from the first K units on the frame - where K is the reciprocal of the sampling fraction - and then selecting every Kth unit in the stratum thereafter.

Note, from the size class definitions, that establishments having zero employees were not included in this survey. Such units were assumed to be out of the scope of the survey.

5. Sample Size Determination and Allocation Within Strata

The total number of establishments selected from the Dun and Bradstreet sampling frame was determined using two stages. The first stage was to compute the target number of respondents for each estimation cell using the standard sample size formula. The required specifying a target relative standard error (RSE) for the cell estimates. The RSE's for this survey were set at the following levels:

Relative standard error (percent)


SIC range:
24 through 29................. 5.0
30 through 39................. 7.5
40 through 80.................10.0

The units were then allocated to size classes within the estimation cells using Neyman allocation. This method allocates based on the number of establishments in each stratum and on the stratum variability in the key design variable (in this case employment). Size class strata having a large number of establishments on the frame or a high variability in employment (as defined by the population variance) received a greater number of sample units than other strata in the sample. Because the larger size classes often have a high variability in employment, this allocation resulted in "oversampling" the larger size classes in a cell. The required number of cases for each stratum are shown in TABLE 1-2 in the column labeled "Target Number of Respondents."


                          TABLE 1-2

Number of Firms, Required Sample Sizes, Calls Made and Completed
________________________________________________________________
                                   Target     Total       Number
 SIC                 Total         Number     Cases    Completed
GROUPS      Size    Plants    Respondents    Called     May 1988
________________________________________________________________
 243        1-19    10,986             39        78           41
           20-99     1,995             32        64           40
          100-249      346             12        24           15
            >250       147             48        98           64
                    ______           ____      ____         ____
           Total    13,474            131       264          160

 245        1-19       864             15        25            9
           20-99       385             15        40           26
          100-249      220             17        45           19
            >250        43             30        43           20
                    ______           ____      ____         ____
           Total     1,512             77       153           74

 249        1-19     4,301             37        74           43
           20-99       888             35        70           45
          100-249      129             11        22           12
            >250        44             12        24           15
                    ______           ____      ____         ____
           Total     5,362             95       190          115

  25        1-19    11,505             20        40           21
           20-99     3,254             26        52           34
          100-249      858             13        26           13
            >250       449             73       146           86
                    ______           ____      ____         ____
           Total    16,066            132       264          154

  26        1-19     3,485             20        44           13
           20-99     2,830             30        62           34
          100-249    1,307             30        54           34
            >250       576            184       384          227
                    ______           ____      ____         ____
           Total     8,198            264       544          308

  27        1-19    64,922             45        60           30
           20-99    10,656             34       122           77
          100-249    1,850             12       126           87
            >250       869             88        50           40
                    ______           ____      ____         ____
           Total    78,297            179       358          234

 281        1-19     1,721             20        52           25
           20-99       735             20        52           33
          100-249      189             20        52           29
            >250       157             96       157           68
                    ______           ____      ____         ____
           Total     2,802            156       313          155

 282        1-19       700             20        40           25
           20-99       499             20        40           25
          100-249      184             20        40           21
            >250       175             58       116           52
                    ______           ____      ____         ____
           Total     1,558            118       236          123

 283        1-19     1,289             25        50           28
           20-99       544             25        50           30
          100-249      179             25        50           26
            >250       205             92       184           92
                    ______           ____      ____         ____
           Total     2,217            167       334          176

 284        1-19     3,065             20        40           23
           20-99       767             20        40           31
          100-249      184             20        40           22
            >250       155             70       140           61
                    ______           ____      ____         ____
           Total     4,171            130       260          137

 285        1-19     1,092             20        50           36
           20-99       549             15        40           20
          100-249      100             10        30           15
            >250        45             37        45           33
                    ______           ____      ____         ____
           Total     1,786             82       165          104

 286        1-19       860             20        54           28
           20-99       346             15        44           29
          100-249       95             15        44           28
            >250        92             50        62           39
                    ______           ____      ____         ____
           Total     1,393            100       204          124

 287        1-19     1,306              8        31           14
           20-99       338              9        33           19
          100-249       57              4        23           17
            >250        43             44        44           21
                    ______           ____      ____         ____
           Total     1,744             65       131           71

 289        1-19     2,562             16        38           26
           20-99       918             23        52           31
          100-249      162              9        24           15
            >250        85             51        85           38
                    ______           ____      ____         ____
           Total     3,727             99       199          110

 291        1-19       606             20        50           23
           20-99       227             20        50           23
          100-249       85             20        50           28
            >250       130             59        90           48
                    ______           ____      ____         ____
           Total     1,048            119       240          122

 295        1-19       862             21        46           27
           20-99       237             26        56           38
          100-249       45             10        24           20
            >250         9              8         9            3
                    ______           ____      ____         ____
           Total     1,153             65       135           88

 299        1-19       516             15        32           24
           20-99       186             22        46           35
          100-249       23              4        11           11
            >250         5              5         5            3
                    ______           ____      ____         ____
           Total       730             46        94           73

 308        1-19     8,062             16        32           18
           20-99     4,249             13        26           22
          100-249    1,162              7        14           17
            >250       388             18        36           23
                    ______           ____      ____         ____
           Total    13,681             54       108           80

  30        1-19     1,983             25        50           31
 (not 308) 20-99       722             25        50           33
          100-249      239             25        50           33
            >250       234             48        96           61
                    ______           ____      ____         ____
           Total     3,178            123       246          158

 311        1-19       283              5        24            3
           20-99       125              9        21           11
          100-249       29              4        16            3
            >250        16              5        10            5
                    ______           ____      ____         ____
           Total       453             23        71           22

  31        1-19     2,232              8        16            6
(not 311)  20-99       610             13        26           20
          100-249      220             13        26           16
            >250       176             13        26           17
                    ______           ____      ____         ____
           Total     3,238             47        94           59

  32        1-19    14,499             15        28           10
           20-99     4,207             34        62           40
          100-249      873             35        64           44
            >250       448             29        53           33
                    ______           ____      ____         ____
           Total    20,027            113       207          127

  33       1-19      4,983             67       201           81
           20-99     2,803             67       201          139
          100-249    1,006             42       126           86
            >250       711             25        75           54
                    ______           ____      ____         ____
           Total     9,503            201       603          360

  34        1-19    29,005             62       113           56
           20-99    11,849            110       200          117
          100-249    2,394             86       157           86
            >250     1,080             62       113           66
                    ______           ____      ____         ____
           Total    44,328            320       583          325

 35, 36,    1-19   117,005            100       200           90
 38, 39    20-99    30,820            126       188          123
          100-249    7,468             93       137           98
            >250     5,657             80       133           84
                   _______           ____      ____         ____
           Total   160,950            399       658          395

 40, 44,    1-19    45,323             20        37           20
  & 45     20-99     5,612             20        37           15
          100-249      799             20        37           15
           > 250       533             50        91           32
                    ______           ____      ____         ____
           Total    52,267            110       202           82

  46        1-19       439             15        28           20
           20-99       162             15        28           21
          100-249       18              8        18           16
            >250         5              5         5            6
                    ______           ____      ____         ____
           Total       624             43        79           63

  49        1-19    12,982             40        73           47
           20-99     4,046             40        73           57
          100-249      844             40        73           58
            >250       558            150       273          206
                    ______           ____      ____         ____
           Total    18,430            270       492          368

50 & 51(1)  1-19    45,422            200       364          233
           20-99     3,464             65       142          106
          100-249      205             30        79           48
            >250        57             57        57           31
                   _______           ____      ____         ____
           Total    49,148            352       642          418

 55 & 75    1-19   284,632             10        30           20
           20-99    20,846             10        30           23
          100-249    1,523             10        30           18
            >250       116             20        60           31
                   _______           ____      ____         ____
           Total   307,117             50       150           92

 72 & 73(2) 1-19   139,889            120       240          119
           20-99     5,511             30        60           33
          100-249      527             20        40           28
            >250       108             25        50           23
                   _______           ____      ____         ____
           Total   146,035            195       390          203

 7641 &     1-19    18,098             60       110           67
 7692      20-99       289             20        48           32
          100-249       12             10         9            8
            >250         1              1         1            0
                    ______           ____      ____         ____
           Total    18,400             91       168          107

  80        1-19   233,984             50        91           50
           20-99    17,174             30        55           34
          100-249    6,310             30        55           39
            >250     3,912            220       400          279
                   _______           ____      ____         ____
           Total   261,380            330       601          402

  37        1-19     9,863             10        19            5
           20-99     2,997             10        19           12
          100-249    1,026             10        19           13
            >250     1,072             70       128           72
                    ______           ____      ____         ____
           Total    14,958            100       185          102
_________________________________________________________________
  Footnote(1): Refers to SIC Codes: 5093, 5193, 5161, 5191, 5198
  Footnote(2): Specifically SIC Codes: 7211, 7213, 7215, 7216, 7218, 7231,
7241, 7251, 7261, 7299, 7732, 7342, and 7395

The number of units actually selected from the D&B frame in each stratum was based on the number of completed cases required for the stratum and on the expected response rate. Almost all sample surveys, especially voluntary surveys, select some number of cases which do not result in a completed interview. In some instances, these will be establishments which have gone out of business, are duplicate cases, or are companies not in the SIC category shown on the frame. Such cases are "Out of Scope". Other establishments, though in scope, refuse to participate or are not reached in the sampling protocol, defined here as a total of five telephone attempts. Experience on surveys similar to the PEL survey indicated that a completion ration of 50-60% was expected for this survey (the ratio of completed questionnaires to total cases which must be drawn - both in and out of scope). However, to be safe, a larger number of cases were selected and held in reserve from the D&B frame so that, if additional sample units needed to be included to reach the target sample sizes, the cases could be easily obtained.

In fact, for the vast majority of cells, a 60 percent completion ratio was realized. The total number of establishments called in each of the sample strata are shown in TABLE 1-2 in the column labeled "Total Cases Called." In general, this number is equal to the target sample divided by 0.60. The number of completed survey responses is shown in TABLE 1-2.

6. Data Collection Methodology

The data collection method chosen for the survey was Computer Assisted Telephone Interviewing (CATI). In this method the interviewer talks to the respondent on the telephone while sitting in front of a computer screen. Each question to be asked appears on the screen in the proper sequence. CATI systems allow for the responses to be examined during the data collection process. Answers that are out of the possible range of responses or which are not consistent with other answers received earlier in the questionnaire can be immediately identified. Another advantage is that it frees the interviewer from using a hard copy questionnaire which requires skipping manually to different parts of the questionnaire based upon the responses. Finally, this method saves resources by creating a machine readable record of the responses at the conclusion of the interview, thereby eliminating the need for keypunching.

7. Establishment Count Comparison

Comparison of the survey establishment count is designed to put into relief both consistencies and inconsistencies between the sample results compared with other existing databases. Table C-4 of the RIA compares the estimated number of establishments from the sample survey with the Dun and Bradstreet (D&B) establishment list (the sample frame), the establishment count from the Bureau of Labor Statistics ES-202 file. In general, the survey produced establishment estimates in between the higher D&B counts and the lower CBP and ES-202 establishment numbers.

8. Variance Estimation

As with any sample survey, quantification of sampling error of estimates is an important function. Errors are quantified by computing the standard error of each estimate produced from the survey. Under certain assumptions, the standard error can be used to make probability statements about estimates. For example, an interval approximately equal to two standard errors on either side of an estimate is a 95 percent of the time, were the survey to be replicated many times.

A replication technique was used to determine standard errors for the PEL survey. Such techniques involve resampling the sample data multiple times to calculate its variability. A replication method was chosen because of two characteristics of the survey. First, some of the estimates which were planned to be produced are nonlinear, such as the benchmarked estimates described above. Second, nonresponse adjustment was used to modify the final weights. In both of these situations, replication-type variance estimators are particularly useful.

The PEL survey was designed using employment as a variability measure. The survey results are consistent with this design; those estimates more closely related to employment had lower relative standard errors. Hence, the RSE for the survey estimate for the total number of production workers over all industries was three percent, for the total number of workers potentially exposed, four percent overall, and for the total numbers of workers overexposed, six percent overall.

Compared to these estimates, the final cost estimates had considerably higher RSE's. This stems from the fact that many establishments were assigned a zero cost, while others in the same stratum were often assigned a very large cost. This "all or none" characteristic of the costing algorithm resulted in an increased RSE for this variable. Even so, for all industries combined, the overall RSE for cost was 11 percent.

9. Treatment of Non-Sampling Errors

An important component to any survey effort is the treatment of nonsampling errors. Examples of such errors are:

* Nonresponse bias - error introduced because some selected respondents either do not respond at all (unit nonresponse) or do not respond to a particular question (item nonresponse);

* Response bias - error introduced due to the way questions are phrased or the way respondents interpret what is being asked (this also includes error due to deliberate misrepresentation of the answers to questions by respondents.

In the PEL survey, the nonresponse problem was dealt with using two standard methodologies. For unit nonresponse, a mean imputation procedure was used. This procedure assumes that there is no fundamental difference between respondents and nonrespondents and, therefore, usable cases can be reweighted to represent the entire universe. For item nonresponse, an imputation scheme which uses related cases in the respondent group to estimate the missing data was used.

The situation for response bias was handled by obtaining information from site visits. OSHA conducted 90 site visits in a cross section of industries. A portion of these visits were performed on establishments which were also in the telephone survey. Data on key variables collected during the telephone survey were compared with information obtained from the site visits. This analysis can be found in Supplement 3.

9.1. Unit nonresponse adjustment

To adjust the sample for those cases selected from the D&B frame which were called but were out of scope (OOS), out of business (OOB), or in scope but unwilling to participate in the survey, the following approach was used.

* All solicited sample units were assigned a response code based on the following categories:

Code and Description

03 Non-working telephone number 04 Incorrect SIC - out of scope 05 Out of Business (OOB) 06 Not a business or wrong business 07 Duplicate record 08 Could not reach respondent after five attempts 09 Communication barrier 10 Initial refusal 11 Mid-interview refusal (did not answer initial chemical and process questions) 12 Completed interview (completed both initial chemical and process questions) 13 Other nonresponse.

* All units having a response code equal to 08, 09, 10, 11, 12, or 13 were classified as viable sample units (in scope, in business). Sample units having a response code equal to 12 were classified as both viable and usable. A nonresponse adjustment weight was assigned to each usable record in the database, based on the ratio of viable to usable sample units in the record's cell and size stratum:

n(1) n(1) NRAF(i)(j)= sum of I(V(k)) / sum of I(U(k)) k = 1 k = 1

where: i = number of estimation cell j = number of the size class I(V(k)) = l if the kth sample unit is viable, =O otherwise; I(U(k)) = l if the kth sample unit is usable, =0 otherwise.

The use of this weight is equivalent to performing a mean imputation for item nonresponse.

The response rate may be defined as the number of usable cases divided by the number of viables, and the completion rate as the number of usables divided by the total number of cases contacted. Using these definitions, the response and completion rates were as follows:

Response rate = 68.7% Completion rate = 60.0%

9.2. Item nonresponse adjustment/impulation

Often survey respondents do not know the answers to some questions or refuse to answer particular questions. In such cases, it is possible to fill in missing values using an imputation scheme. The idea is to use information from both the respondent (answers to other questions which they did supply) and information from other respondents (those answering the missing question) in order to estimate a reasonable response to the missing datum.

The imputation method chosen for the PEL survey is a hybrid method which combines the concepts of a mean imputation and a "hot-deck" imputation. A mean imputation method replaces the missing values on a certain question with the mean value from those respondents answering that question. A hot deck method attempts to find a respondent who matches the respondent having a missing value (in terms of other survey characteristics) and uses the value of the "twin" to replace the missing value. The method used here is a hybrid in the sense that it employs a mean imputation, but only over a small segment of the population which obviously matches the respondent having a missing value.

In particular, the procedure examines three or four digit SIC subgroups within the estimation cell by size class. The mean values of the responses to a particular question of interest in such sample subgroupings were used to impute the missing values in that grouping. In the case of categorical variables (for example, YES/NO questions), a randomization scheme was used randomly supplied the appropriate set of responses to missing questions based on a probability distribution determined from those who responded.

It should be noted that the values which were placed on the database were not intended to be estimates of the missing responses. Rather, they are meant to be substitute responses which allow the case to be used in the generation of survey estimates. In the aggregate, estimates produced using imputed data make sense for use in aggregate estimates, but may not be useful for the individual establishment. Care was taken in the imputation program to be sure that imputed responses were consistent with other answers for the establishment of interest. Original responses to all questions were retained on the sample record and all responses representing imputed values were identified.

One set of questions which was not imputed for was whether monitoring for the presence of certain toxic chemicals was done at the establishment. the data collected produced an estimate, for those establishments where chemicals or processes were found, that 15.8 percent did monitoring, 71.9 percent did not do monitoring, and 12.3 percent of respondents did not know or refused to answer the question. Of those establishments that did monitor, 25.6 percent provided the requested data.

10. Survey Instrument

As mentioned earlier, data collection for PEL survey was accomplished by Computer Assisted Telephone Interviewing. Prior to calling, a letter was sent to each selected establishment. This letter is shown in Exhibit 1-1. Also, a hard copy version of the PEL questionnaire is given in Exhibit 1-2.


                         EXHIBIT 1-1

U.S. Department of Labor
                                   Assistant Secretary for
                                   Occupational Safety and Health
                                   Washington D.C. 20210

                                   SIC Code 3479
                                   Metal Coating & Allied Serv.
February 25, 1988                  OMB Approval No. 1218-0142

Mr. John Q. Sample
Chairman
Anycompany
123 Sample St.
Anytown, US 12345

Dear Mr. Sample:

The Occupational Safety and Health Administration (OSHA) of the U.S.
Department of Labor is required by law to set permissible exposure limits
for chemical substances in the workplace.  Current exposure limits were
set 17 years ago using values established by the American Conference
of Governmental Industrial Hygienists (ACGIH) and the American National
Standards Institute (ANSI).

OSHA has begun a process for revising out-of-date permissible exposure
limits.  To ensure that any new exposure limits take into account actual
workplace conditions, we are conducting a voluntary survey of U.S.
business establishments.  Included will be questions about specific
processes which we believe are performed in your industry and a limited
(no more than 10 per process) list of chemicals which we believe are
involved in those processes.  Your facility was selected to be included in
the study.

Decisions regarding new permissible exposure limits will be improved
significantly if we have input from as many firms as possible. the
interview will take about 30 minutes.  Names of responding firms will not
be associated with their answers, and all data will be treated as
confidential by our contractor.

Please help us expedite the survey process by returning to us, within one
week, the enclosed postage paid card with the name and phone number of the
person in your organization our contract interviewer should contact.  If
this card is not received, a representative of our contractor, KCA
Research, Inc., will call your office directly to conduct the interview or
be directed to the company official designated by you.

Enclosed is a list of the topic areas for the survey.  This may help in
preparing for the interview.

We appreciate your cooperation and look forward to receiving the
information we need from your designated representative.

Sincerely,

John A. Pendergrass
Assistant Secretary for OSHA

Enclosure:

        TOPICS COVERED BY SURVEY

I.  GENERAL FIRM CHARACTERISTICS

  * Primary activity at this location
  * Approximate numbers of production & maintenance workers
  * Number of shifts per day and length of shift

II.  IDENTIFICATION OF GENERAL PROCESSES PERFORMED BY FIRM

  * Chemicals used in specific processes or operations and estimated
quantities involved
  * Approximate a number of work stations or assembly lines used and number
of workers at each
  * Description of process engineering controls such as ventilation and
enclosure
  * Estimated frequency of process or operation performance
  * Description of personal protective equipment used, including
respirators, eye, face, and skin protection
  * Information regarding exposure monitoring

                        EXHIBIT 1-2

  ________, we are conducting a survey on behalf of OSHA to assess the
current practices of all types of businesses in the handling of toxic and
hazardous chemicals.  A letter was sent to you informing you of this
survey.

1.  Did you receive our letter?
    1 = Yes
    2 = No

If answer "Yes", begin next paragraph with "As you know,"
If answer "No", begin next paragraph with "I'm sorry.  Let me summarize
what the letter said about the survey".

  We are interested in understanding all significant operations or
processes in your firm that generate dust, mist, fumes, gases or vapor
that your employees might potentially encounter.  Of course, all responses
and trade or technological secrets will be kept strictly confidential and
no company-specific information will be released to OSHA.

2.  Should I direct my questions to you, or is there someone else in the
firm who you feel would be better qualified to answer?

    1 = Yes, this person will answer survey
    2 = No, call: Name_______________
                  Title______________
                  Phone______________
    C = Call back (Set up time for recontact)
    R = Refused to answer (Terminate interview)
    D = Don't Know/ No Response

Let me begin by asking some general questions about your facility

3.  Our records show your firm to be engaged in__________.?
Is this correct?  (Interviewer will read title or brief description for
this SIC code.)

    1 = Yes
    2 = No, our function her is____________________
    C
    R
    D

4.  How many production workers do you have at this location?

    1 = ____________ production workers
    C = Call back
    R = Refused to answer
    D = Don't know

5.  How many maintenance workers (for example: painters, welders & cleaning staff) do you employ?

    1 = ____________ maintenance workers
    2 = Production workers do maintenance functions
    3 = None, only clerical, managerial, or sales personnel
    C
    R
    D

5a. Of these maintenance workers, how many do painting as their primary
work activity?

    1 = ____________ do painting as primary activity
    2 = None
    C
    R
    D

5b. Of these maintenance workers, how many do welding as their primary
work activity?

    1 = ____________ do welding as primary activity
    2 = None
    C
    R
    D

6.  How many shifts per day (24 hr. period) do you have at this location?

    1 = ____________ shifts/24 hr.
    C
    R
    D

I now want to ask you some questions about chemicals which we believe are
common among firms in your industry.  (These chemicals would be selected
on the basis of large volume usage, known toxicity, or know exposure
problems in excess of permissible limits as identified from NOES or IMIS
or from industry expert opinion).

7.  which of the following chemicals are used,processed, or emitted at
your facility?

    Chemical A       1 = Yes     2 = No     C      R     D

(The interviewer will read chemical list specified for this 4-digit SIC.
If "Don't Know" (D) is the response, the interviewer will then attempt to
clarify the question by reading a list of common synonyms for the
chemical.  The subsequent answer can then be reassessed as "Yes" or "No".

8.  Are there any other chemicals in major use in your operations that I
did not list?

    1 = Yes (Skip to #8 and add to list)
    2 = No
    C
    R
    D

9.  What is the approximate quantity of chemical A that your facility
purchases each week or month?

    1 = ______ lbs. per week purchased
    2 = ______ gals. per week purchased
    3 = ______ lbs. per month purchased
    4 = ______ gals. per month purchased
    C
    R
    D

Repeat Question #9 until all identified chemicals are quantified.

10.  Have exposure limits been adopted by your firm for these chemicals?

    1 = Yes
    2 = No
    C    )
    R    ) (Skip to #12)
    D    )

11.  What exposure limits have been adopted?

    1 = OSHA PEL's
    2 = NIOSH REL's
    3 = ACGIH TLV's
    4 = Other ____________
    C
    R
    D

The next questions are about processes/operations which we believe are
common among firms in your industry

12.  Are any of the following processes/operations performed in your
facility?

Operation #1         1 = Yes     2 = No     C     R     D

(Interviewer would read list of up to 6 processes or operations specified
for this 4-digit SIC code.  This list would be identified from secondary
data sources and industry experts.  If information regarding relevant
processes was not available or sufficient, then this question would be
rephrased to elicit process/operation identification from the respondent.)

13.  Are there any other processes/operations at your facility that I did
not list?

    1 = Yes  (Skip to #12 and add to list)
    2 = No
    C
    R
    D

For each identified process/operation, ask questions 14 - 26

14.  In Process/Operation 1:

     Is Chemical A used?    1 = Yes     2 = No     C     R     D

REPEAT UNTIL ALL IDENTIFIED CHEMICALS HAVE BEEN ASKED ABOUT USAGE IN THIS
PROCESS

15.  How many work stations (or assembly lines) are involved in this
process/operation?

    1 = ______ work stations
    2 = ______ assembly lines
    C
    R
    D

16.  On average, how many workers are directly involved in this
process/operation at each work station (or assembly line)?

    1 = ______ workers/work station
    2 = ______ workers/assembly line
    C
    R
    D

17.  Of these workers, what percent work exclusively at this
process/operation?

    1 = 100%  (Go to #18)
    2 = ___%
    C        )
    R        )(Go to #18)
    D        )

17a. For those workers who do not work exclusively at this
process/operation, in what other processes/operations are they also
employed?

    1 = ___________________________
    C
    R
    D

18.  Is this process/operation a completely enclosed activity?

    1 = Yes   (Skip to #14)
    2 = No
    C
    R
    D

19.  Is this process/operation located outdoors?

    1 = Yes   (Skip to #21)
    2 = No
    C
    R
    D

20.  Is this process/operation ventilated?

    1 = Yes
    2 = No  )
    C       ) (Skip to #21)
    R       )
    D       )

20a. What is the type of ventilation?

    1 = Local exhaust     1 = Yes     2 = No     C     R     D
    2 = General dilution
    3 = Natural ventilation
    4 = Other (specify type) ___________________

21.  How often is this process/operation performed during each shift?

    1 = Continuously over entire shift, every shift
    2 = Daily (specify # / day)________
    3 = Weekly (specify # / week)_______
    4 = Monthly (specify # / month)______
    5 = Yearly (specify # / year)________
    6 = Other (specify # / period)_______
    C
    R
    D

22.  Are respirators routinely used by workers?

    1 = Yes
    2 = No  )
    C       ) (Skip to #23)
    R       )
    D       )

22a.  What type of respirator?

    1 = Single use
    2 = Half-Mask cartridge
    3 = Half-Mask canister
    4 = Full-face cartridge
    5 = Full-face canister
    6 = Powered air purifying respirator
    7 = Air supplied respirator
    8 = self-contained breathing apparatus
    9 = Escape respirator
    10= Other ___________________________
    C
    R
    D

23. Do you provide maintenance workers who have exposure to this process
with respirators?

    1 = Yes
    2 = No
    C
    R
    D

24. Is skin, face, or eye protection used?

    1 = Yes
    2 = No  )  (Skip #25)
    C       )
    R       )
    D       )

24a. What type(s) of skin, face, or eye protection?

    1 = Long sleeve shirt
    2 = Coverall
    3 = Apron
    4 = Gloves
    5 = Chemical Protective Clothing
    6 = Goggles
    7 = Face Shield
    8 = Other ______________________
    C
    R
    D

25. Do you have a hazard communications training program for these
workers?

    1 = Yes
    2 = No
    C
    R
    D

26. Has environmental monitoring been done at or near this
process/operation?

    1 = Yes
    2 = No  )
    C       )  (Skip to #14 until all processes surveyed)
    R       )
    D       )

26a. Has this monitoring been designed to evaluate control of:

    1 = potential short term (less than 15 min.) exposures?  (STEL)
    2 = potential 15 minute - 4 hour exposures?
    3 = potential 4 - 8 hour exposures? (TWA)
    C
    R
    D

26b. During this monitoring, were any chemicals found to be in excess of
your adopted exposure guidelines?

    1 = Yes
    2 = No )
    C      )  (Skip to #27)
    R      )
    D      )

26c. Which chemical(s) were found to exceed adopted guidelines?

    1 = __________________
    C       )
    R       )  (Skip to #27)
    D       )

26d. What activity, work process or operation do you feel is most
responsible for the exposures above your adopted guidelines?

    1 = ____________________
    2 = Not able to specify
    C
    R
    D

27. Can you give us your monitoring data for process 1?

    1 = Yes
    2 = No  )
    C       )  (Skip to #14)
    R       )
    D       )

27a. What is the name of the first (next) chemical for which you have
monitoring data?

    1 = ______________________
    C
    R
    D

27b. Is the data based on actual monitoring readings or is it estimated?

    1 = Actual
    2 = Estimated
    C
    R
    D

27c. Is the data for the work area or for the person (worker)?

    1 = Area
    2 = Person
    C
    R
    D

27d. Is the data recorded for the individual worker or the work process?

    1 = Worker
    2 = Process
    C
    R
    D

27f. Is the unit of measurement parts per million or milligrams per cubic
meter?

    1 = PPM
    2 = Mg/M(3)
    C
    R
    D

27g. What is the exposure data for this chemical?

    1 = __________________
    C
    R
    D

27h. Do you have exposure estimates for other chemicals used in this
process?

    1 = Yes   (Skip to #27a)
    2 = No  )
    C       ) (Skip to #14 until all processes surveyed)
    R       )
    D       )

28. What do you estimate to be the market value of plant and equipment at
your facility?

    1 = Less than $50,000
    2 = $50,000 - $500,000
    3 = $501,000 - $1,000,000
    4 = $1 - 5 million
    5 = $5 - 50 million
    6 = More than $50 million
    C
    R
    D

29. Can you estimate the annual value of shipments from your facility?

    1 = Less than $50,000
    2 = $50,000 - $500,000
    3 = $500,000 - $1,000,000
    4 = $1 - 5 million
    5 = $5 - 50 million
    6 = More than $50 million
    C
    R
    D

Thank you for cooperating with us in our survey.

[54 FR 2332, Jan. 19, 1989; 54 FR 28054, July 5, 1989; 54 FR 28154, July 5, 1989]


Regulations (Preambles to Final Rules) - Table of Contents Regulations (Preambles to Final Rules) - Table of Contents